2023 年的期刊文章 · 第 6 页

共 1511 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2023-10-13 到 2023-10-01。

按期刊、按主题看 →

第 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 页

JRSSB Other 讨论 / 评论 · 发表于 2023-10-13

David Siegmund's contribution to the Discussion of “Estimating means of bounded random variables by betting” by Waudby-Smith and Ramdas

David Siegmund

摘要

This contribution by Waudby-Smith and Ramdas provides a host of new insights combined with a concise, yet comprehensive, historical review of “any time valid inference”, which has blossomed in recent years after the hiatus that followed its beginnings in the mid-20th Century. Particularly, attractive is the authors’ systematic use of martingales of the form (29) to facilitate the betting interpretation and give unified derivations of a large diversity of inequalities. Given the success of this paper in studying the mean of a distribution, one might ask how to estimate the entire distribution. A special case of Kelly’s gambling theory involves a histogram having N cells with probabilities p1,…,pN⁠, which pays a constant amount N when an observation falls into the nth cell for n=1,…,N⁠. The gambler’s expected log return is maximised by repeatedly betting for each n a fraction pn of the gambler’s current fortune on the nth cell; and the procedure can be effectively simulated by a statistician who does not know the cell probabilities, but judiciously estimates them sequentially (cf. Robbins & Siegmund, 1974). Can one schedule a sequence of refinements of the histograms to approximate a continuous probability density function and estimate the density efficiently? A second exercise is to give a sequence of anytime valid confidence bands for a distribution function. A hint in this direction occurs in Siegmund (1988), but one might want to use an estimator that pays more attention to the tails of the distribution. The subject of anytime valid inference disappeared from the literature for 40-some years after an initial burst of theoretical activity, presumably because applications were lacking. The huge amount of data available in some internet experiments changes that situation, but a genuine scientific application would, in my view, be an important contribution. My favourite candidate is sequential clinical trials, where experimental expense and government regulations usually mandate tests having a fixed horizon. Is there evidence that the flexibility of anytime valid stopping rules to deal with changed horizons outweighs their disadvantages? (I found disappointing what I was able to show 40 years ago.) *(abstract via OpenAlex)*

SMMR Nonparametric & Semiparametric Genetics & Omics · 发表于 2023-10-13

A novel rare variants association test for binary traits in family-based designs via copulas

Houssou R. G. Dossa, Alexandre Bureau, Michel Maziade, Lajmi Lakhal-Chaieb, Karim Oualkacha

摘要

With the cost-effectiveness technology in whole-genome sequencing, more sophisticated statistical methods for testing genetic association with both rare and common variants are being investigated to identify the genetic variation between individuals. Several methods which group variants, also called gene-based approaches, are developed. For instance, advanced extensions of the sequence kernel association test, which is a widely used variant-set test, have been proposed for unrelated samples and extended for family data. Family data have been shown to be powerful when analyzing rare variants. However, most of such methods capture familial relatedness using a random effect component within the generalized linear mixed model framework. Therefore, there is a need to develop unified and flexible methods to study the association between a set of genetic variants and a trait, especially for a binary outcome. Copulas are multivariate distribution functions with uniform margins on the [Formula: see text] interval and they provide suitable models to capture familial dependence structure. In this work, we propose a flexible family-based association test for both rare and common variants in the presence of binary traits. The method, termed novel rare variant association test (NRVAT), uses a marginal logistic model and a Gaussian Copula. The latter is employed to model the dependence between relatives. An analytic score-type test is derived. Through simulations, we show that our method can achieve greater power than existing approaches. The proposed model is applied to investigate the association between schizophrenia and bipolar disorder in a family-based cohort consisting of 17 extended families from Eastern Quebec.

SiM Survival Analysis Causal Inference Clinical Trials · 发表于 2023-10-12

The survival‐incorporated median vs the median in the survivors or in the always‐survivors: What are we measuring? and Why?

Qingyan Xiang, Ronald J. Bosch, Judith J. Lok

摘要

Many clinical studies evaluate the benefit of a treatment based on both survival and other continuous/ordinal clinical outcomes, such as quality of life scores. In these studies, when subjects die before the follow‐up assessment, the clinical outcomes become undefined and are truncated by death. Treating outcomes as “missing” or “censored” due to death can be misleading for treatment effect evaluation. We show that if we use the median in the survivors or in the always‐survivors as estimands to summarize clinical outcomes, we may conclude that a trade‐off exists between the probability of survival and good clinical outcomes, even in settings where both the probability of survival and the probability of any good clinical outcome are better for one treatment. Therefore, we advocate not always treating death as a mechanism through which clinical outcomes are missing, but rather as part of the outcome measure. To account for the survival status, we describe the survival‐incorporated median as an alternative summary measure for outcomes in the presence of death. The survival‐incorporated median is the threshold such that 50% of the population is alive with an outcome above that threshold. Through conceptual examples and an application to a prostate cancer treatment study, we show that the survival‐incorporated median provides a simple and useful summary measure to inform clinical practice.

SMMR Bayesian Methods High-Dimensional · 发表于 2023-10-12

SIGHR: Side information guided high-dimensional regression

Yuan Yang, Christopher S McMahan, Yu-Bo Wang, James W Baurley, Sung-Shim Park

摘要

In this work, we develop a novel Bayesian regression framework that can be used to complete variable selection in high dimensional settings. Unlike existing techniques, the proposed approach can leverage side information to inform about the sparsity structure of the regression coefficients. This is accomplished by replacing the usual inclusion probability in the spike and slab prior with a binary regression model which assimilates this extra source of information. To facilitate model fitting, a computationally efficient and easy to implement Markov chain Monte Carlo posterior sampling algorithm is developed via carefully chosen priors and data augmentation steps. The finite sample performance of our methodology is assessed through numerical simulations, and we further illustrate our approach by using it to identify genetic markers associated with the nicotine metabolite ratio; a key biological marker associated with nicotine dependence and smoking cessation treatment.

SMMR High-Dimensional Survival Analysis · 发表于 2023-10-12

Penalized variable selection in multi-parameter regression survival modeling

Fatima-Zahra Jaouimaa, Il Do Ha, Kevin Burke

摘要

Standard survival models such as the proportional hazards model contain a single regression component, corresponding to the scale of the hazard. In contrast, we consider the so-called “multi-parameter regression” approach whereby covariates enter the model through multiple distributional parameters simultaneously, for example, scale and shape parameters. This approach has previously been shown to achieve flexibility with relatively low model complexity. However, beyond a stepwise type selection method, variable selection methods are underdeveloped in the multi-parameter regression survival modeling setting. Therefore, we propose penalized multi-parameter regression estimation procedures using the following penalties: least absolute shrinkage and selection operator, smoothly clipped absolute deviation, and adaptive least absolute shrinkage and selection operator. We compare these procedures using extensive simulation studies and an application to data from an observational lung cancer study; the Weibull multi-parameter regression model is used throughout as a running example.

JRSSB Machine Learning Testing & Uncertainty 讨论 / 评论 · 发表于 2023-10-11

Philip S Thomas, Erik Learned-Miller and My Phan's contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Philip S Thomas, Erik Learned-Miller, My Phan

摘要

We enthusiastically agree about the importance of developing tighter confidence intervals and sequences for the mean—the topic of this paper. There are increasing concerns about the safety and fairness of machine learning systems, and we believe one powerful tool for ensuring that machine learning systems are safe, fair, and otherwise reliable is for them to provide guarantees in the form of their confidence that the models they produce will satisfy certain safety or fairness guarantees. At the core of such methods are mechanisms for producing confidence intervals on parameters related to fairness and safety. However, often these confidence intervals must be constructed from samples from extreme distributions for which normality assumptions remain unreasonable even with relatively large numbers of samples. In such cases, confidence intervals like those proposed in this paper are critical, because they do not rely on normality assumptions yet remain relatively tight. We appreciate the comparison of the Hedged-CI method to our own recent confidence interval for the mean (‘PTL21’ in the paper), and find the further increase in tightness compelling, as it could result in improved data-efficiency of safe and fair machine learning methods. Furthermore, we find the extension of the proposed methods to the confidence sequence setting compelling, as it could allow for the re-testing of the safety or fairness of machine learning models as additional data becomes available. One additional comparison that would be interesting to see would be to the confidence interval proposed by Gaffke (2005). (Among the three statistics presented by Gaffke, we are referring to statistic ‘K’.) Showing that this confidence interval has (or does not have) guaranteed coverage remains an open question. Nevertheless, given the remarkable tightness of Gaffke’s bound and the substantial effort that has gone into finding distributions for which it does not exhibit coverage (with no success), it seems like a promising direction for future analysis. *(abstract via OpenAlex)*

JRSSB Testing & Uncertainty Optimization & Learning Theory 讨论 / 评论 · 发表于 2023-10-11

Philip B. Stark’s contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Philip B Stark

摘要

This paper is a gem, and I predict it will have lasting impact on theoretical and applied statistics. Drawing on intellectual heritage spanning more than 80 years, illustrating and leveraging deep connections among results in probability, finance, information theory, computer science, and statistics, Waudby-Smith and Ramdas present mature, intuitive, flexible, computable, and powerful methods for two fundamental (indeed, canonical) nonparametric inference problems: conservative confidence bounds for the expected value of a bounded random variable from IID observations and conservative confidence bounds for the mean of a finite, bounded population from a simple random sample. Sequentially valid methods have been proposed for those problems and the related problem of conservative tests for nonnegative or bounded means [including some based on Ville’s (1939) inequality, explicitly or through Wald’s (1945) sequential probability ratio test: Howard et al., 2021; Kaplan, 1987, 2012; Stark, 2009, 2020, 2023; Stark & Teague, 2014], but their performance varies and their derivations provide little insight into how to sharpen the methods. In contrast, the betting martingale framework leads naturally to techniques that approximately optimise power, are computationally tractable, perform well ‘out of the box’ for a broad range of bounded populations and distributions, and can incorporate prior information without compromising coverage. Moreover, the sequential validity makes them ‘safe’ against common practices that invalidate other approaches (optional stopping and optional continuation) (Grünwald et al., 2023; Ramdas et al., 2021; Shafer, 2021). It would be interesting to compare these new methods to the approach of Orabona and Jun (2021) based on regret bounds for a universal portfolio and to the conjectured bound of Gaffke (2005) and Learned-Miller and Thomas (2019) for fixed sample size, even though the latter bound has not been proved to be valid in general. Accommodating Bernoulli sampling is trivial (see, e.g. Ottoboni et al., 2020; Stark, 2023), but tied to the question of derandomisation, especially for fixed sample size. [Kaplan (1987) does this in a crude way using the union bound; Waudby-Smith and Ramdas give a number of sharper approaches, and work in progress makes further improvements; Ramdas & Manole, 2023.] Accommodating stratified sampling using union–of–intersection tests is straightforward (Spertus et al., 2023; Spertus & Stark, 2022; Stark, 2023) because independent E-values can be combined by multiplication or averaging (Vovk & Wang, 2021); alternatively, the E-values can be converted to P-values and then combined. Optimising union–of–intersection tests based on test martingales yields an interesting generalisation of the multi-arm bandit (MAB) problem, a ‘gang of bandits’, where N MABs are probed in parallel: at each stage, the statistician chooses an arm, which pulls the same arm of all N MABs, as if they were ‘ganged’ like switches. The statistician has a fortune for each MAB and bets separately on each, but seeks to maximise their minimum fortune across the bandits (Stark, 2023) by choosing the sequence of arms and bets. Improved methods for stratified inference have immediate applications ranging from election audits (Ottoboni et al., 2020; Spertus & Stark, 2022; Stark, 2008, 2009, 2020, 2023; Stark & Teague, 2014; Waudby-Smith et al., 2021) and financial audits to randomised experiments with blocking. In closing, I highly recommend appendix F for a brief history of connections among gambling, portfolio theory, martingales, concentration inequalities, and hypothesis testing. *(abstract via OpenAlex)*

JRSSB Other 讨论 / 评论 · 发表于 2023-10-11

Anthony C Davison and Igor Rodionov’s contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Anthony C Davison, Igor Rodionov

摘要

We congratulate the authors on a wide-ranging contribution to a statistical hardy perennial. The restriction that the support be known and bounded appears highly restrictive, yet Proposition 3 suggests much wider potential for applications, one of which we briefly explore below. The Pickands–Balkema–de Haan theorem (Balkema & de Haan, 1974; Pickands, 1975) establishes that the distribution of a random variable X, conditioned on its exceeding a high threshold u can in rather wide generality be approximated by the generalised Pareto distribution (GPD). This result, a cornerstone of the statistical analysis of rare events, is typically applied by choosing a threshold u empirically from a random sample of n observations, fitting the GPD to the k observations that exceed u, and using this fit to estimate quantiles or other measures of risk. Numerous procedures have been suggested to choose u, or equivalently k, often based on threshold-stability properties of the GPD; informal graphical approaches were proposed by Davison and Smith (1990), and in many settings more formal procedures are complemented by these or other graphs. The Hill plot (Hill, 1975) is used when X lies in the Fréchet max-domain of attraction with tail index 1/γ and is related to the Rényi representation for exponential order statistics. One potential use of the ideas in Section 4.4 of the paper is to aid in the choice of k, using the sample order statistics X(1)≤⋯≤X(n) and the simple martingale both (Mk) and its expectation 1/(1+γ) when the scaled differences k(logX(n−k+1)−logX(n−k)) are independent exponential variables lie in the unit interval. This leads to a conservative overall approach to choosing the number of upper order statistics k used in the Hill estimator. Figure 1 shows confidence intervals for γ for a log-normal sample and for the Danish fire insurance data (Embrechts et al., 1997). The first uses λk+=(γ+1)/4 and λk−=(γ+1)/(4γ)⁠, and the second uses the λs suggested in expressions (25) and (26) of the paper. The theory does not apply to the first, for which γ=0⁠, but extremes of finite log-normal samples are typically well-approximated by a distribution with a heavier tail; the martingale shows the expected lack of stability. The second is more stable, suggesting that γ≈0.75⁠; the corresponding confidence interval is similar to that in the top left-hand panel of Figure 6.4.3 of Embrechts et al. (1997). Hill plots showing the estimate of γ (solid) and martingale 95% confidence limits (dotted) based on a log-normal sample (left) and the Danish insurance data (right). It should be clear that we found the paper very stimulating. *(abstract via OpenAlex)*

JRSSB Testing & Uncertainty 讨论 / 评论 · 发表于 2023-10-10

Steven R Howard's contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Steven R Howard

摘要

I congratulate the authors on this creative and well-written contribution. It’s always a delight to see new insights into such a fundamental problem as inference for the mean of bounded observations, and certainly inspires excitement about the future of game-theoretic statistical inference. Besides the impressive technical contributions, I appreciate how the authors draw previously unappreciated connections between work in information theory, finance, and online learning, hopefully increasing future collaboration between researchers in these fields. In their conclusion, the authors allude to the possibility of estimating functionals other than the mean, using the quantile as an example. Are there fruitful applications of betting methods to estimation of general, possibly vector-valued functionals θ defined by possibly vector-valued estimating equations of the form E[ψ(Xt,θ)|X1,…,Xt−1]=0 a.s. for all t, assuming such a value of θ exists (Angelopoulos et al., 2023)? This would be interesting even for univariate functionals beyond the mean, though the real prize would be practical methods for estimating vector-valued functionals such as population regression coefficients, for which we have few tractable results (such as the multivariate normal mixture (Abbasi-Yadkori et al., 2011), Hermitian dilation (Tropp, 2012), and Banach space methods (Pinelis, 1994)). Martingale methods have found adoption in practical applications, and martingale-based confidence sequences specifically have found use in some online A/B testing systems, especially in platforms designed for use by general audience with little statistical training (Johari et al., 2022; Maharaj et al., 2023). However, there remains resistance due to that fact that fixed-sample confidence intervals based on the CLT, and group sequential methods which are generally supported by CLT arguments, yield tighter intervals and higher power. Of course, CLT-based confidence intervals do not come with a nonasymptotic guarantee, and can seriously break in cases of practical interest, so we should not be surprised that nonasymptotic intervals are wider than those implied by the CLT. However, for example, even in the ideal case of Bernoulli(1/2) observations, no matter how large the sample size, Hoeffding’s bound remains wider than the CLT interval by a factor of 2log(2/α)/z1−α/2⁠, where zp is the p-quantile of the standard normal distribution. This factor is about 1.4 at α=0.95⁠, implying an experimenter must gather roughly twice as many samples to obtain the same interval width with a Hoeffding bound that she would have achieved with the CLT; this can be a hard sell. Ideally, a nonasymptotic confidence interval Cn with width |Cn| might have a guarantee like the following: for i.i.d. (Xi) with variance σ2⁠, we have n|Cn|→\,p2z1−α/2σ as n→∞⁠. On the other hand, a linear sub-Gaussian uniform boundary is tight for Brownian motion, suggesting that martingale-based methods are ideal for extremely frequent ‘peeking’ but not for fixed-sample testing or sparse ‘peeking’. We are left wondering whether nonasymptotic analysis based on martingales and betting can be made similarly competitive in the latter regime. Lastly, the authors discuss in Supplementary Material Section E.5 a simulation with non-i.i.d. data, but the framework still excludes nonstationary settings in which the conditional mean changes with time. This is in contrast to the exponential supermartingale approach which gives a confidence sequence for a sequence of possibly random ‘estimands’ given by the average conditional means θt=t−1∑i=1tE(Xi|X1,…,Xi−1) (Howard et al., 2021). Such robustness to general nonstationary is reassuring in general and useful in specific cases such as the estimation of average treatment effects in a potential outcomes model. Do we have counterexamples showing that betting-based confidence sequences, derived under the assumption of constant conditional mean, substantially break down in this more general setting? Is it possible that the bounds in this paper are valid in the more general nonstationary setting and simply await a new analysis? Supplementary material is available online at Journal of the Royal Statistical Society: Series B. *(abstract via OpenAlex)*

SiM Longitudinal Data Nonparametric & Semiparametric Bayesian Methods · 发表于 2023-10-10

Semiparametric normal transformation joint model of multivariate longitudinal and bivariate time‐to‐event data

An‐Ming Tang, Cheng Peng, Niansheng Tang

摘要

Joint models for longitudinal and survival data (JMLSs) are widely used to investigate the relationship between longitudinal and survival data in clinical trials in recent years. But, the existing studies mainly focus on independent survival data. In many clinical trials, survival data may be bivariately correlated. To this end, this paper proposes a novel JMLS accommodating multivariate longitudinal and bivariate correlated time‐to‐event data. Nonparametric marginal survival hazard functions are transformed to bivariate normal random variables. Bayesian penalized splines are employed to approximate unknown baseline hazard functions. Incorporating the Metropolis‐Hastings algorithm into the Gibbs sampler, we develop a Bayesian adaptive Lasso method to simultaneously estimate parameters and baseline hazard functions, and select important predictors in the considered JMLS. Simulation studies and an example taken from the International Breast Cancer Study Group are used to illustrate the proposed methodologies.

JRSSB Other 讨论 / 评论 · 发表于 2023-10-10

Ruodu Wang's contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Ruodu Wang

摘要

I congratulate Ian Waudby-Smith and Aaditya Ramdas for this excellent contribution to the theory of e-testing (or testing by betting). Over the past several years, this theory has been developed rapidly in multiple exciting directions. The contribution of Waudby-Smith and Ramdas is an important piece in this active stream of literature. I have a few comments on the approach in the current paper and its generalisations. The main approach taken in the paper (see its Theorem 1), which is standard in the field of game-theoretical statistics (Shafer & Vovk, 2019), relies on constructing an e-process (Mtθ)t∈T for each parameter value θ∈Θ (which is the mean m in the paper). The e-process (Mtθ)t∈T is often (but not always) constructed by combining several sequential e-variables Eθ=(Etθ)t∈T from the data (i.e. satisfying Eθ[Etθ|Ft−1]≤1 for each t∈T and θ∈Θ⁠), via a method of martingale: Mtθ=∏s=1t(1−λsθ(Esθ−1)),t∈T, where λ=(λtθ)t∈T is a predictable process. The abstract problem of combining sequential e-variables is studied in Vovk and Wang (2022), where it is shown that the above martingale method is the only admissible way to combine sequential e-variables into one e-variable. Therefore, anytime validity (i.e. validity under optional sampling) is obtained automatically if the goal is to make a decision based on a combined e-value. Although the above method of e-testing is by now standard and its validity is easy to show, the highly non-trivial tasks are to build suitable Eθ and to find powerful λθ⁠. The validity is guaranteed even when the data-generating procedure varies arbitrarily over time, as long as the parameter of interest (mean m in this paper) in the null hypothesis is specified. Nevertheless, the power and optimality of λθ depend crucially on how data are generated. Most methods (such as GRAPA and aGRAPA introduced in the current paper) use sample mean, sample variance, or the empirical distribution to decide λθ⁠. This requires some ‘stationarity’, ‘predictability’, or ‘temporal structure’ of the data. In some applications involving dynamic decision making (data depend on previous decisions), such as backtesting financial risk prediction, such stationarity cannot be assumed. In this context, some options of powerful betting strategies are studied by Wang et al. (2022). The e-testing approach can be applied to many other quantities in a model-free fashion, similar to the mean with bounded supported treated in this paper. Wang et al. (2022) developed one-sided e-tests for other quantities, including mean (with one-side bounded support), variance, quantile, and the risk measure Expected Shortfall. A main advantage of such methods is that they do not assume any knowledge of the data-generating probability or its temporal structure. *(abstract via OpenAlex)*

JRSSB Testing & Uncertainty 讨论 / 评论 · 发表于 2023-10-10

Jiayi Li, Yuantong Li and Xiaowu Dai's contribution to the Discussion of ‘Estimating means of bounded random variables by betting' by Waudby-Smith and Ramdas

Jiayi Li, Yuantong Li, Xiaowu Dai

摘要

We congratulate Waudby-Smith and Ramdas for their interesting paper Waudby-Smith and Ramdas (2020b) in generating confidence intervals and time-uniform confidence sequences for mean estimation with bounded observations. Their methodology utilises composite non-negative martingales and establishes a connection to game-theoretic probability. Our comments will focus on numerical comparisons with alternative methods. The corresponding code is available at https://github.com/Likelyt/Estimate-mean-with-betting. Consider a sequence of random variables (Xt)t=1∞⁠, drawn from a distribution P∈Pμ⁠, where Pμ represents the set of all distributions on [0,1]∞⁠. We assume EP[Xt|X1,…,Xt−1]=μ for some unknown μ∈[0,1]⁠. The objective is to construct a time-uniform confidence sequence (Ct)t=1∞ that satisfies the condition The confidence sequence guarantees that, for any fixed n, Cn is a (1−α)-confidence interval for μ. We review three methods for generating such confidence sequences. The predictable plug-in empirical Bernstein (Pr-EB) method, discussed in Section 3 of Waudby-Smith and Ramdas (2020b), combines the Robbins’ method of mixture (Robbins, 1970) with exponential supermartingales. The hedged capital process (Betting) method, introduced in Section 4 of Waudby-Smith and Ramdas (2020b), is a novel approach that enjoys the interpretation of wealth accumulation in a game and has connections to the game-theoretic probability (Shafer & Vovk, 2005). The Bootstrap resampling method is implemented using the R package Boot (Canty & Ripley, 2022). With B=200 bootstrap replicates and L=10 batches, we calculate a separate confidence interval for data sequence 2l≤t<2l+1 within each batch l=1,…,L⁠. The confidence intervals are constructed using the α2L- and (1−α2L)-quantiles of the bootstrap means. We sequentially generate data from Beta(10,30) distribution for 1≤t≤104⁠, and construct a 95% confidence sequence Ct in equation (1) using methods (a)–(c) in Section 1. Figure 1 shows that the Betting method outperforms the Bootstrap method with a higher lower bound in the confidence sequence and consistently tighter intervals for t≥1,500⁠. This aligns with the intuition that the Bootstrap method results in wider intervals due to dividing the confidence budget among data subsets when dealing with a large number of sequential data points. Moreover, the Betting method outperforms the Pr-EB method with consistently tighter confidence sequences for t≤3,000⁠. For t≥3,000⁠, both methods yield comparable coverage lengths. The Betting method’s sequence shows a slight shift towards smaller values, reflecting the left-skewed ground truth distribution, while the Pr-EB method produces a symmetric interval around the estimated mean. Comparisons based on the synthetic data of Beta distribution. (a) Confidence sequence and (b) coverage length. We analyse a batting dataset of 18 Major League players from the 1970 season, available in Efron and Hastie (2021) or the R package EfronMorris. The goal is to construct a 95% confidence interval for the true batting level of each player based on their first 45 at-bats. The following results are obtained from 100 data replications. Figure 2a shows the average confidence intervals, where the black circle represents the true batting level of each player. Note that while the Bootstrap method fails to cover the true batting level of player 17, the Betting method successfully includes it. Figure 2b presents the coverage probability of both methods. The Betting method achieves higher coverage probability for the true batting levels of players 4 and 17 compared to the Bootstrap method. These results indicate that the Betting method outperforms the Bootstrap method in constructing accurate confidence intervals for this baseball data. Comparisons based on the baseball batting data from Efron and Hastie (2021). (a) Confidence interval and (b) coverage probability. The work in Waudby-Smith and Ramdas (2020b) could inspire several directions for future research. One of such directions is the generalisation of the Betting method to handle multidimensional observations. Currently, the Betting method relies on the assumption that the underlying capital process is a martingale. However, when estimating the mean of multidimensional data, applying Ville’s inequality (Waudby-Smith & Ramdas, 2020a) for confident rejection of the null hypothesis becomes more challenging. Defining the hedged capital process in a vector space introduces complexities when simultaneously estimating multiple attributes. Another direction is extending the Betting method to online decision-making scenarios, such as dynamic treatment, online recommendation, online matching, and dynamic pricing (Dai & Jordan, 2021a, 2021b; Kamenica & Gentzkow, 2011; Mansour et al., 2020). These tasks frequently involve constructing confidence sequences to determine optimal actions. It is of interest to study the applications of the Betting method to these contexts. *(abstract via OpenAlex)*

JRSSB Other 讨论 / 评论 · 发表于 2023-10-10

Hien Nguyen’s contribution to the Discussion of “Estimating means of bounded random variables by betting” by Waudby-Smith and Ramdas

Hien Nguyen

摘要

Hien Nguyen; Hien Nguyen's Contribution to the Discussion of “Estimating means of bounded random variables by betting” by Waudby-Smith and Ramdas, Journal of th *(abstract via OpenAlex)*

JRSSB Other 讨论 / 评论 · 发表于 2023-10-09

Vladimir Vovk's contribution to the Discussion of “Estimating means of bounded random variables by betting” by Waudby-Smith and Ramdas

Vladimir Vovk

摘要

Vladimir Vovk; Vladimir Vovk's contribution to the Discussion of “Estimating means of bounded random variables by betting” by Ian Waudby-Smith and Aaditya Ramda *(abstract via OpenAlex)*

JRSSB Testing & Uncertainty 讨论 / 评论 · 发表于 2023-10-09

Seconder of the vote of thanks to Waudby-Smith and Ramdas and contribution to the Discussion of ‘Estimating means of bounded random variables by betting’

Gergely Neu

摘要

The paper proposes an elegant new framework for deriving time-uniform confidence intervals for the mean of a sequence of random variables. The proposed technique is based on a game-theoretic view of statistical inference, rephrasing the problem of building confidence intervals as identifying a set of coin-betting games with a certain ‘plausibility’ property: if one can get ‘implausibly rich’ in a game by betting on the mean being lower (resp. higher) than a certain threshold, then the mean must be higher (resp. lower) than said threshold. A confidence interval can then be derived as the interval between the lowest and highest values that cannot be ruled out as ‘implausible’. This game-theoretic view extends the pioneering work of Ville (1939) and Shafer and Vovk (2019) who used similar betting games to lay an alternative set of foundations of probability theory that is free from measure theory. The authors of this work develop a set of new concentration inequalities from this framework that are both aesthetically pleasing and empirically tight. In many ways, the framework appears to be so natural and profound that the reader wonders why this connection with betting has not been exploited in this depth previously. Indeed, the language of betting is so closely tied with the theory of test martingales that one wonders why almost all past work on deriving confidence intervals (and more generally, testing) have focused on using supermartingales that inevitably lead to at least some looseness when used for this purpose. In other words, I believe that the authors have contributed a valuable fundamental insight which will undoubtedly lead to many interesting future results. Besides this appraisal, I would also like to mention some limitations of the results presented in the paper and particularly highlight connections with the concurrent work of Orabona and Jun (2021) that proposes a similar framework for deriving confidence sequences. One slightly unsatisfying aspect of the paper is that, while the authors develop a new theory of mean estimation through the lens of playing a sequential game, they do not seem to make full advantage of this connection by going ‘all in’ on this connection. In particular, most of the proposed estimators are based on betting heuristics rather than betting strategies with already existing theoretical guarantees on their growth rate (or other properties of interest). There are several such algorithms for a range of known betting games, and I am surprised to see that these are only mentioned as an afterthought in the appendix (if at all). In particular, it remains unclear what betting algorithm one should use to, say, minimise the confidence width. The authors criticise the use of principled betting strategies, saying that those are all optimised for worst-case price relatives, and the i.i.d. case considered in the paper should be much easier to handle. On one hand, this is hardly an appropriate justification to replace principled methods with heuristics. On the other hand, this argument is not entirely well-founded, as several well-studied betting strategies are ‘equalisers’ in the sense that they have the same regret for all sequences (i.e. worst-case or i.i.d.). See the detailed discussion on this issue by Orabona and Jun (2021). The confidence intervals depend on the order in which the data are presented to the algorithm. This is natural for non-i.i.d. data but is hard to justify when the data are i.i.d. The authors propose a ‘derandomisation’ strategy consisting of shuffling the data a number of times, executing the algorithm on all resulting sequences, and aggregating the results. Besides being somewhat unsatisfactory from a conceptual point of view, this approach also destroys the efficiency that was part of the magic of the proposed method. I wish to point out that this aspect has been addressed very neatly by Orabona and Jun (2021), whose reduction to the universal portfolio optimisation method of Cover and Ordentlich (1996) produces estimates that are independent of the order of the data without such derandomisation. The method is limited to random variables bounded almost surely in [0,1]⁠. This seems like an inherent limitation, although it seems likely that it can be removed either by combining the current approach with the ‘moment truncation’ estimator of Catoni (2012) (as one of the authors did in a related paper, Wang & Ramdas, 2023). Still, it is unclear if such extensions can retain the appealing ‘purity’ of the approach proposed in the present paper, or if a more elegant and natural treatment of unbounded random variables would be possible. I wish to emphasise that these latter comments are not meant to diminish the value of the contributions of the paper but rather to highlight some outstanding issues that future work has to address. There are of course several more open questions left behind that this note can hardly do justice to. That said, opening so many potential directions for future research is a sign of great fundamental work, and I wish to congratulate the authors (as well as the concurrent pioneers; Orabona & Jun, 2021) for bringing this set of tools to the forefront of statistical inference. The vote of thanks was passed by acclamation. *(abstract via OpenAlex)*

JRSSB Testing & Uncertainty 讨论 / 评论 · 发表于 2023-10-09

Ryan Martin’s contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Ryan Martin

摘要

Congratulations to Waudby-Smith and Ramdas (WSR) for their excellent contribution to the rapidly growing literature on anytime-valid inference. Since some of my work involves imprecise probability, Professor Ramdas asked me privately what, if any, connections there are between e-values, etc. and imprecise probability. The answer to his question might be of general interest, so I’ll share it here. Following WSR, let (Xt:t≥1) be a [0,1]-valued process with distribution P∈Pμ having unknown mean μ. Write Xt=(X1,…,Xt) and xt for a generic realisation. Take M(Xt,m) to be an e-value for testing H0m:μ=m⁠. Given xt⁠, define This function determines an imprecise probability for uncertainty quantification about μ, one whose upper probability is a sub-additive possibility measure given by Γ¯xt is a coherent upper probability, so it determines a nonempty, closed, and convex credal set C(Γ¯xt)={Q:Q(⋅)≤Γ¯xt(⋅)} of probability distributions dominated by Γ¯xt⁠. The credal set facilitates probabilistic uncertainty quantification about μ: a hypothesis H is discredited if Q(H) is small for all Q∈C(Γ¯xt)⁠, i.e. if Γ¯xt(H) is small. These are not Bayesian-style subjective degrees of belief about μ—they’re frequentistly justified. Indeed, WSR’s confidence sets “Ct” are exactly the γ-level sets: Ct≡{m:γxt(m)>α}⁠, and there’s a uniform validity property (e.g. Cella & Martin, 2023), that prevents systematic discreditation of any true hypothesis about μ. But there’s another possibility measure that’s even better than Γ¯xt⁠. Define If M is such that supm∈[0,1]πxt(m)=1 for all xt⁠, then Π¯xt(A)=supm∈Aπxt(m) is a possibility measure. It can be shown (Martin, 2022a, 2022b) that Π¯Xt enjoys the same properties as Γ¯Xt⁠, but is more efficient, i.e. πXt(⋅)≤γXt(⋅) and, hence, C(Π¯Xt)⊆C(Γ¯Xt)⁠. For WSR’s nonparametric case, computation of πxt in (1) is a challenge, so one might opt for the computationally simpler but statistically less efficient upper bound γxt⁠. In other cases, however, more can be done. For example, in the context of parametric models, I argued (Martin, 2023) that one can efficiently balance frequentist desiderata with the likelihood principle by excluding those unrealistic and impractically extreme stopping rules, e.g. sample until the test rejects. This exclusion is achieved by suitably shrinking the domain over which an optimisation like in (1) is taken. *(abstract via OpenAlex)*

JRSSB Testing & Uncertainty 讨论 / 评论 · 发表于 2023-10-09

Rong Jiang and Keming Yu's contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Rong Jiang, Keming Yu

摘要

We want to congratulate the authors on estimating means of bounded random variables in both with- and without-replacement settings. The authors constructed confidence intervals and time-uniform confidence sequences for the mean of a bounded random variable using test supermartingale technique. It deepens our understanding of the confidence sequence. Confidence sequence is one particular tool in sequential design that facilitates anytime-valid inference. In particular, confidence sequence is a sequence of confidence intervals that is valid at data-dependent stopping times. We offer three comments. First, the CtPrPl−EB in Theorem 2 involves (λt)t=1∞⁠. The authors recommend the predictable plug-in (λtPrPl−EB)t=1∞ given by Whether so many parameter estimators λtPrPl−EB will lead to the superposition of errors and the failure of the method. The Hoeffding process (MtH(m))t=0∞ in equation (8) only need one λ. In particular, when t<100⁠, is CtPrPl−EB still correct? See Figure 2, we can see the results are bed when t<100⁠. Thus, whether Theorem 2 needs to add restrictions on t. Moreover, whether CtPrPl−EB is sensitive to c in λtPrPl−EB⁠, and if so, how to select c, although 1/2 or 3/4 is recommended. Second, the author mentioned their test supermartingle can also be inverted to get a confidence sequence for any quantile. How about mode (Chernoff, 1964), expectile (Newey & Powell, 1987), and extremile (Daouia et al., 2019)? Third, the authors consider arbitrary distribution but bounded distribution which implies all moments exist and require a Chernoff-type assumption on the distribution resulting in O(logt/t) shrinkage rates for the confidence sequences. Recently, Wang and Ramdas (2023) show that employing Catoni’s estimator improves the rate to O(loglog2t/t) under weaker assumptions on the distribution ((1+δ)-th moment bound). We wonder if the methods and results can be generalised to unbounded observations, since the σ2-bounded-variance assumption (Wang & Ramdas, 2023) is more realistic and easier to verify. *(abstract via OpenAlex)*

JASA Networks & Graphs · 发表于 2023-10-09

Network Estimation by Mixing: Adaptivity and More

Tianxi Li, Can M. Le

摘要

Networks analysis has been commonly used to study the interactions between units of complex systems. One problem of particular interest is learning the network’s underlying connection pattern given a single and noisy instantiation. While many methods have been proposed to address this problem in recent years, they usually assume that the true model belongs to a known class, which is not verifiable in most real-world applications. Consequently, network modeling based on these methods either suffers from model misspecification or relies on additional model selection procedures that are not well understood in theory and can potentially be unstable in practice. To address this difficulty, we propose a mixing strategy that leverages available arbitrary models to improve their individual performances. The proposed method is computationally efficient and almost tuning-free for network modeling. We show that the proposed method performs equally well as the oracle estimate when the true model is included as individual candidates. More importantly, the method remains robust and outperforms all current estimates even when the models are misspecified. Extensive simulation examples are used to verify the advantage of the proposed mixing method. Evaluation of link prediction performance on more than 500 real-world networks from different domains also demonstrates the universal competitiveness of the mixing method across multiple domains. Supplementary materials for this article are available online. *(abstract via OpenAlex)*

JRSSB Other 讨论 / 评论 · 发表于 2023-10-09

Martin Larsson and Johannes Ruf’s contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Martin Larsson, Johannes Ruf

摘要

We congratulate Ian Waudby-Smith and Aaditya Ramdas on their comprehensive and insightful paper. The authors construct time-uniform confidence sequences for the mean of a sequence of [0,1]-valued random variables X1,X2,… that all have the same conditional mean, assumed to be deterministic. Specifically, fix m∈(0,1) and define Pm as the set of probability measures on the canonical sequence space under which, for each t∈N⁠, the conditional expectation of Xt given the history X1,…,Xt−1 is equal to the deterministic number m⁠. The authors construct nonnegative processes (Kt) that satisfy K0=1 and are Pm-martingales, i.e. P-martingales for each P∈Pm⁠. These martingales are then used to construct anytime-valid statistical tests that in turn can be transformed into confidence sequences (see also Ramdas et al., 2022). In keeping with the game-theoretic probability literature, the authors refer to the processes (Kt) as capital processes. Let us ponder this terminology, starting with the following simple but interesting observation made by the authors: every nonnegative Pm-martingale (Kt) with K0=1 is of the form for some predictable process (λt) with values in [−(1−m)−1,m−1]⁠. Predictable means that each λt only depends on X1,…,Xt−1⁠. One may interpret λt as the proportion of one’s capital Kt−1 that is invested in an asset with return Xt−m⁠, keeping whatever is left over ‘in the pocket’. The fact that λt can be greater than one, or negative, poses no issue as this simply means that one may borrow cash to purchase more of the asset than one could otherwise afford, or sell the asset short to generate additional cash income. Crucially, one’s capital must always remain nonnegative. The upshot is this: not only is (Kt) the capital process produced by repeated betting; thanks to the representation (1) there is always an explicit trading strategy, operating on one single asset, that generates the capital process. Indeed, one has λt=(Kt/Kt−1−1)/(Xt−m)⁠. Given any particular (Kt) of interest, we believe insight can be gained by computing the associated trading strategy. For example, the ‘diversified Kelly’ capital process considered by the authors is built from D separate strategies (λtd)⁠, d=1,…,D⁠. This is equivalent to the single strategy where (Ktd) is the capital process generated by the dth strategy. In other words, diversified Kelly arises from executing the capital-weighted average of the given strategies. This links it to Cover’s universal portfolios (Cover, 1991). Many of the capital processes proposed in the paper are specified in terms of a strategy (λt)⁠. What we find worth emphasising is that such a (λt) can always be found, and is likely to yield insights. Finally, let us point out that the representation (1) is of course specific to the particular structure of Pm⁠. An interesting question is to what extent analogous representations exist for other, more complex, statistical hypotheses. *(abstract via OpenAlex)*

JRSSB Machine Learning 讨论 / 评论 · 发表于 2023-10-09

Anastasios N. Angelopoulos’ contribution to the Discussion of ‘Estimating means of bounded random variables by betting’ by Waudby-Smith and Ramdas

Anastasios N Angelopoulos

摘要

I congratulate Waudby-Smith and Ramdas for their paper, which develops algorithms for confidence sequences motivated by betting. An impressive consequence of their work is the derivation of confidence intervals for a fixed sample size n, given by Theorem 3/Remark 3 and Theorem 4/Remark 4. These bounds, which my colleagues and I have referred to eponymously as the WSR bounds, prove quite useful for quantifying the uncertainty of artificial intelligence algorithms. I will give an example application from our work en route towards showcasing their utility. In order to quantify and guarantee the reliability of a machine learning model on a sample point X, we may seek to form a risk-controlling prediction set (RCPS) T(X) (Angelopoulos et al., 2021; Bates et al., 2021). More formally, given a test data point (X,Y) and a calibration dataset Dcal⁠, we want to calibrate the model to produce sets that control some [0,1]-bounded loss L in high probability: The loss quantifies the reliability of the set; an example is the false-negative proportion L(T(x),y)=|y∩T(x)|/|y| in the case that y itself is set-valued. Critically, we seek to satisfy (1) in a distribution-free way, and in finite samples, by concentrating on the empirical risk. The central point of my commentary is to highlight the significance of the WSR inequality in the formation of an RCPS. Any concentration inequality will work to construct an RCPS, but the tightness of the bound greatly affects the usability of the prediction sets. In our practical experiments, the WSR bound outperformed all other concentration inequalities, such as those of Hoeffding (1994), Bentkus (2004), and Maurer and Pontil (2009), and furthermore, was nearly tight, adapting quite strongly to the variance of the loss. Figure 1 depicts a practical application of the WSR bound, demonstrating its use in segmenting gut tumours through machine learning. Here, we focus on controlling the false-negative rate—essentially, the proportion of the tumour that is incorrectly excluded from the segmentation mask. This is of crucial importance in ensuring that virtually all cancerous cells are excised with a high degree of certainty. The application of the WSR bound to tumour segmentation. True positives are white, true negatives are black, false positives are blue, and false negatives are red. Comparing the WSR’s practical improvements over Hoeffding’s inequality, the difference is clear, and the bound is tight for all practical purposes. This has been evident in our works on risk control and machine learning-assisted inference (Angelopoulos et al., 2023). I extend my gratitude to the authors for this practical contribution. The following contributions were received in writing after the meeting *(abstract via OpenAlex)*

SiM Bayesian Methods Latent Variables & Mixtures Spatial & Time Series · 发表于 2023-10-09

A spatio‐temporal Dirichlet process mixture model for coronavirus disease‐19

Jaewoo Park, Seorim Yi, Won Chang, Jorge Mateu

摘要

Understanding the spatio‐temporal patterns of the coronavirus disease 2019 (COVID‐19) is essential to construct public health interventions. Spatially referenced data can provide richer opportunities to understand the mechanism of the disease spread compared to the more often encountered aggregated count data. We propose a spatio‐temporal Dirichlet process mixture model to analyze confirmed cases of COVID‐19 in an urban environment. Our method can detect unobserved cluster centers of the epidemics, and estimate the space‐time range of the clusters that are useful to construct a warning system. Furthermore, our model can measure the impact of different types of landmarks in the city, which provides an intuitive explanation of disease spreading sources from different time points. To efficiently capture the temporal dynamics of the disease patterns, we employ a sequential approach that uses the posterior distribution of the parameters for the previous time step as the prior information for the current time step. This approach enables us to incorporate time dependence into our model in a computationally efficient manner without complicating the model structure. We also develop a model assessment by comparing the data with theoretical densities, and outline the goodness‐of‐fit of our fitted model.

JASA Bayesian Methods · 发表于 2023-10-09

A Nonstationary Soft Partitioned Gaussian Process Model via Random Spanning Trees

Zhao Tang Luo, Huiyan Sang, Bani Mallick

摘要

There has been a long-standing challenge in developing locally stationary Gaussian process models concerning how to obtain flexible partitions and make predictions near boundaries. In this work, we develop a new class of locally stationary stochastic processes, where local partitions are modeled by a soft partition process via predictive random spanning trees that leads to highly flexible spatially contiguous subregion shapes. This valid nonstationary process model knits together local models such that both parameter estimation and prediction can be performed under a unified and coherent framework, and it captures both discontinuities/abrupt changes and local smoothness in a spatial random field. We propose a theoretical framework to study the Bayesian posterior concentration concerning the behavior of this Bayesian nonstationary process model. The performance of the proposed model is illustrated with simulation studies and real data analysis of precipitation rates over the contiguous United States. Supplementary materials for this article are available online. *(abstract via OpenAlex)*

SiM Bayesian Methods High-Dimensional Genetics & Omics · 发表于 2023-10-08

Genome‐wide search algorithms for identifying dynamic gene co‐expression via Bayesian variable selection

Wenda Zhang, Zichen Ma, Lianming Wang, Daping Fan, Yen‐Yi Ho

摘要

A wealth of gene expression data generated by high‐throughput techniques provides exciting opportunities for studying gene‐gene interactions systematically. Gene‐gene interactions in a biological system are tightly regulated and are often highly dynamic. The interactions can change flexibly under various internal cellular signals or external stimuli. Previous studies have developed statistical methods to examine these dynamic changes in gene‐gene interactions. However, due to the massive number of possible gene combinations that need to be considered in a typical genomic dataset, intensive computation is a common challenge for exploring gene‐gene interactions. On the other hand, oftentimes only a small proportion of gene combinations exhibit dynamic co‐expression changes. To solve this problem, we propose Bayesian variable selection approaches based on spike‐and‐slab priors. The proposed algorithms reduce the computational intensity by focusing on identifying subsets of promising gene combinations in the search space. We also adopt a Bayesian multiple hypothesis testing procedure to identify strong dynamic gene co‐expression changes. Simulation studies are performed to compare the proposed approaches with existing exhaustive search heuristics. We demonstrate the implementation of our proposed approach to study the association between gene co‐expression patterns and overall survival using the RNA‐sequencing dataset from The Cancer Genome Atlas breast cancer BRCA‐US project.

SiM Other · 发表于 2023-10-08

Estimation of non‐monotonic transition rates in a semi‐Markov process with covariates adjustments and application to caregivers' stress data

Esther Ngan, Wenyaw Chan, Luis Leon‐Novelo, Valory Pavlik

摘要

With the large ongoing number of aged people and Alzheimer's disease (AD) patients worldwide, unpaid caregivers have become the primary sources of their daily caregiving. Alzheimer's family caregivers often suffer from physical and mental morbidities owing to various reasons. The aims of this paper were to develop alternate methods to understand the transition properties, the dynamic change, and the long‐run behavior of AD caregivers' stress levels, by assuming their transition to the next level only depends on the duration of the current stress level. In this paper, we modeled the transition rates in the semi‐Markov Process with log‐logistic hazard functions. We assumed the transition rates were non‐monotonic over time and the scale of transition rates depended on covariates. We also extended the uniform accelerated expansion to calculate the long‐run probability distribution of stress levels while adjusting for multiple covariates. The proposed methods were evaluated through an empirical study. The application results showed that all the transition rates of caregivers' stress levels were right skewed. Care recipients' baseline age was significantly associated with the transitions. The long‐run probability of severe state was slightly higher, implying a prolonged recovery time for severe stress patients.

SMMR Clinical Trials · 发表于 2023-10-06

Optimal study designs for cluster randomised trials: An overview of methods and results

Samuel I Watson, Alan Girling, Karla Hemming

摘要

There are multiple possible cluster randomised trial designs that vary in when the clusters cross between control and intervention states, when observations are made within clusters, and how many observations are made at each time point. Identifying the most efficient study design is complex though, owing to the correlation between observations within clusters and over time. In this article, we present a review of statistical and computational methods for identifying optimal cluster randomised trial designs. We also adapt methods from the experimental design literature for experimental designs with correlated observations to the cluster trial context. We identify three broad classes of methods: using exact formulae for the treatment effect estimator variance for specific models to derive algorithms or weights for cluster sequences; generalised methods for estimating weights for experimental units; and, combinatorial optimisation algorithms to select an optimal subset of experimental units. We also discuss methods for rounding experimental weights, extensions to non-Gaussian models, and robust optimality. We present results from multiple cluster trial examples that compare the different methods, including determination of the optimal allocation of clusters across a set of cluster sequences and selecting the optimal number of single observations to make in each cluster-period for both Gaussian and non-Gaussian models, and including exchangeable and exponential decay covariance structures.

Biometrics Genetics & Omics · 发表于 2023-10-05

Multiresolution Categorical Regression for Interpretable Cell-Type Annotation

Aaron J. Molstad, Keshav Motwani

摘要

In many categorical response regression applications, the response categories admit a multiresolution structure. That is, subsets of the response categories may naturally be combined into coarser response categories. In such applications, practitioners are often interested in estimating the resolution at which a predictor affects the response category probabilities. In this paper, we propose a method for fitting the multinomial logistic regression model in high dimensions that addresses this problem in a unified and data-driven way. Our method allows practitioners to identify which predictors distinguish between coarse categories but not fine categories, which predictors distinguish between fine categories, and which predictors are irrelevant. For model fitting, we propose a scalable algorithm that can be applied when the coarse categories are defined by either overlapping or nonoverlapping sets of fine categories. Statistical properties of our method reveal that it can take advantage of this multiresolution structure in a way existing estimators cannot. We use our method to model cell-type probabilities as a function of a cell's gene expression profile (i.e., cell-type annotation). Our fitted model provides novel biological insights which may be useful for future automated and manual cell-type annotation methodology.

JASA Causal Inference · 发表于 2023-10-03

Unveiling the Unobservable: Causal Inference on Multiple Derived Outcomes

Yumou Qiu, Jiarui Sun, Xiao-Hua Zhou

摘要

In many applications, the interest is in treatment effects on random quantities of subjects, where those random quantities are not directly observable but can be estimated based on data from each subject. In this article, we propose a general framework for conducting causal inference in a hierarchical data generation setting. The identifiability of causal parameters of interest is shown under a condition on the biasedness of subject level estimates and an ignorability condition on the treatment assignment. Estimation of the treatment effects is constructed by inverse propensity score weighting on the estimated subject level parameters. A multiple testing procedure able to control the false discovery proportion is proposed to identify the nonzero treatment effects. Theoretical results are developed to investigate the proposed procedure, and numerical simulations are carried out to evaluate its empirical performance. A case study of medication effects on brain functional connectivity of patients with Autism spectrum disorder (ASD) using fMRI data is conducted to demonstrate the utility of the proposed method. Supplementary materials for this article are available online. *(abstract via OpenAlex)*

SiM Other · 发表于 2023-10-03

Simultaneous selection and incorporation of consistent external aggregate information

Yunxiang Huang, Chiung‐Yu Huang, Mi‐Ok Kim

摘要

Interest has grown in synthesizing participant level data of a study with relevant external aggregate information. Several efficient and flexible procedures have been developed under the assumption that the internal study and the external sources concern the same population. This homogeneity condition, albeit commonly being imposed, is hard to check due to limitedly available external information in aggregate data forms. Bias may be introduced when the assumption is violated. In this article, we propose a penalized likelihood approach that avoids undesirable bias by simultaneously selecting and synthesizing consistent external aggregate information. The proposed approach provides a general framework which incorporate consistent external information from heterogeneous study populations as long as the conditional distribution of the dependent variable under investigation is same and differences in the independent variable distributions are properly accounted for via a semi‐parametric density ratio model. The proposed approach also properly accounts for the sampling errors in the external information. A two‐step estimator and an optimization algorithm are proposed for computation. We establish the selection and estimation consistency and the asymptotic normality of the two‐step estimator. The proposed approach is illustrated with an analysis of gestational weight gain management studies.

Biometrika Nonparametric & Semiparametric · 发表于 2023-10-03

On geometric convergence for the Metropolis-adjusted Langevin algorithm under simple conditions

Alain Oliviero-Durmus, Éric Moulines

摘要

Summary: While the Metropolis-adjusted Langevin algorithm is a popular and widely used Markov chain Monte Carlo method, very few papers derive conditions that ensure its convergence. In particular, to the authors’ knowledge, assumptions that are both easy to verify and guarantee geometric convergence, are still missing. In this work, we establish V-uniformly geometric convergence for the Metropolis-adjusted Langevin algorithm under mild assumptions about the target distribution. Unlike previous work, we only consider tail and smoothness conditions for the potential associated with the target distribution. These conditions are quite common in the Markov chain Monte Carlo literature. Finally, we pay special attention to the dependence of the bounds we derive on the step size of the Euler–Maruyama discretization, which corresponds to the proposed Markov kernel of the Metropolis-adjusted Langevin algorithm.

SiM Latent Variables & Mixtures Longitudinal Data · 发表于 2023-10-03

Joint clustering multiple longitudinal features: A comparison of methods and software packages with practical guidance

Zihang Lu, Mojtaba Ahmadiankalati, Zhiwen Tan

摘要

Clustering longitudinal features is a common goal in medical studies to identify distinct disease developmental trajectories. Compared to clustering a single longitudinal feature, integrating multiple longitudinal features allows additional information to be incorporated into the clustering process, which may reveal co‐existing longitudinal patterns and generate deeper biological insight. Despite its increasing importance and popularity, there is limited practical guidance for implementing cluster analysis approaches for multiple longitudinal features and evaluating their comparative performance in medical datasets. In this paper, we provide an overview of several commonly used approaches to clustering multiple longitudinal features, with an emphasis on application and implementation through R software. These methods can be broadly categorized into two categories, namely model‐based (including frequentist and Bayesian) approaches and algorithm‐based approaches. To evaluate their performance, we compare these approaches using real‐life and simulated datasets. These results provide practical guidance to applied researchers who are interested in applying these approaches for clustering multiple longitudinal features. Recommendations for applied researchers and suggestions for future research in this area are also discussed.

JASA Genetics & Omics Latent Variables & Mixtures · 发表于 2023-10-03

Exact Decoding of a Sequentially Markov Coalescent Model in Genetics

Caleb Ki, Jonathan Terhorst

摘要

In statistical genetics, the sequentially Markov coalescent (SMC) is an important family of models for approximating the distribution of genetic variation data under complex evolutionary models. Methods based on SMC are widely used in genetics and evolutionary biology, with significant applications to genotype phasing and imputation, recombination rate estimation, and inferring population history. SMC allows for likelihood-based inference using hidden Markov models (HMMs), where the latent variable represents a genealogy. Because genealogies are continuous, while HMMs are discrete, SMC requires discretizing the space of trees in a way that is awkward and creates bias. In this work, we propose a method that circumvents this requirement, enabling SMC-based inference to be performed in the natural setting of a continuous state space. We derive fast, exact procedures for frequentist and Bayesian inference using SMC. Compared to existing methods, ours requires minimal user intervention or parameter tuning, no numerical optimization or E-M, and is faster and more accurate. *(abstract via PubMed)*

JASA Causal Inference · 发表于 2023-10-03

Estimation and Inference of Extremal Quantile Treatment Effects for Heavy-Tailed Distributions

David Deuber, Jinzhou Li, Sebastian Engelke, Marloes H. Maathuis

摘要

Causal inference for extreme events has many potential applications in fields such as climate science, medicine, and economics. We study the extremal quantile treatment effect of a binary treatment on a continuous, heavy-tailed outcome. Existing methods are limited to the case where the quantile of interest is within the range of the observations. For applications in risk assessment, however, the most relevant cases relate to extremal quantiles that go beyond the data range. We introduce an estimator of the extremal quantile treatment effect that relies on asymptotic tail approximation, and use a new causal Hill estimator for the extreme value indices of potential outcome distributions. We establish asymptotic normality of the estimators and propose a consistent variance estimator to achieve valid statistical inference. We illustrate the performance of our method in simulation studies, and apply it to a real dataset to estimate the extremal quantile treatment effect of college education on wage. Supplementary materials for this article are available online. *(abstract via OpenAlex)*

JASA Causal Inference Testing & Uncertainty · 发表于 2023-10-03

A Negative Correlation Strategy for Bracketing in Difference-in-Differences

Ting Ye, Luke Keele, Raiden Hasegawa, Dylan S. Small

摘要

The method of difference-in-differences (DID) is widely used to study the causal effect of policy interventions in observational studies. DID employs a before and after comparison of the treated and control units to remove bias due to time-invariant unmeasured confounders under the parallel trends assumption. Estimates from DID, however, will be biased if the outcomes for the treated and control units evolve differently in the absence of treatment, namely if the parallel trends assumption is violated. We propose a general identification strategy that leverages two groups of control units whose outcomes relative to the treated units exhibit a negative correlation, and achieves partial identification of the average treatment effect for the treated. The identified set is of a union bounds form that involves the minimum and maximum operators, which makes the canonical bootstrap generally inconsistent and naive methods overly conservative. By using the directional inconsistency of the bootstrap distribution, we develop a novel bootstrap method to construct confidence intervals for the identified set and parameter of interest when the identified set is of a union bounds form, and we theoretically establish the asymptotic validity of the proposed method. We develop a simple falsification test and sensitivity analysis. We apply the proposed strategy for bracketing to study whether minimum wage laws affect employment levels. Supplementary materials for this article are available online. *(abstract via OpenAlex)*

SiM Testing & Uncertainty Diagnostics & Prediction · 发表于 2023-10-02

A new accuracy metric under three classes when subclasses are involved and its confidence interval estimation

Nan Nan, Lili Tian

摘要

Compound multi‐class classification” refers to the setting where three or more main classes are involved and at least one of the main classes have multiple subclasses. A common practice in evaluating biomarker performance under “compound multi‐class classification” is “subclasses pooling.” In this article, we first explore the downsides of accuracy metrics based on pooled data. Then we propose a new accuracy measure proper for “compound multi‐class classification” with three ordinal main classes, namely “volume under compound surface ().” The proposed evaluates the accuracy of a biomarker appropriately by identifying main classes without requiring specification of an ordering for marker values of subclasses within each main class. For confidence interval estimation of , both parametric and nonparametric methods are studied, and simulation studies are carried out to assess coverage probabilities. A subset of Alzheimer's Disease Neuroimaging Initiative study dataset is analyzed.

AoS Other · 发表于 2023-10-01

The impacts of unobserved covariates on covariate-adaptive randomized experiments

Yang Liu, Feifang Hu

摘要

Covariate-adaptive randomization (CAR) is commonly implemented in clinical trials to balance observed covariates. Recent studies have demonstrated the advantages of CAR procedures in balancing covariates and improving the subsequent statistical analysis. Covariate balance is crucial, but it is not a panacea for the valid statistical inferences. If the response to a treatment interacts with some unobserved covariates, the conclusion drawn from a CAR experiment may be affected, and thus, be inconsistent with other evidence. This paper aims to demonstrate the relationships between unobserved covariates and the analysis of treatment and covariate effects in CAR experiments. We first derive the asymptotic properties of the statistical methods based on a linear model framework with interactions between the treatment and an unobserved covariate. We also provide sufficient conditions for the identifiability of the treatment and covariate effects. Our results theoretically explain how inconsistent estimations are generated in CAR experiments when some important covariates are unobserved. Under these sufficient conditions, we show that the tests for the treatment and covariate effects can have reduced Type I errors under CAR procedures. A residual-based adjusted test is proposed to recover the Type I error when the effect can be correctly estimated. Numerical studies are conducted to evaluate the performance of our proposed procedure and theoretical findings.

AoS High-Dimensional Testing & Uncertainty · 发表于 2023-10-01

The Lasso with general Gaussian designs with applications to hypothesis testing

Michael Celentano, Andrea Montanari, Yuting Wei

摘要

The Lasso is a method for high-dimensional regression, which is now commonly used when the number of covariates p is of the same order or larger than the number of observations n. Classical asymptotic normality theory does not apply to this model due to two fundamental reasons: (1) The regularized risk is nonsmooth; (2) The distance between the estimator θˆ and the true parameters vector θ∗ cannot be neglected. As a consequence, standard perturbative arguments that are the traditional basis for asymptotic normality fail. On the other hand, the Lasso estimator can be precisely characterized in the regime in which both n and p are large and n/p is of order one. This characterization was first obtained in the case of Gaussian designs with i.i.d. covariates: here we generalize it to Gaussian correlated designs with non-singular covariance structure. This is expressed in terms of a simpler “fixed-design” model. We establish nonasymptotic bounds on the distance between the distribution of various quantities in the two models, which hold uniformly over signals θ∗ in a suitable sparsity class and over values of the regularization parameter. As an application, we study the distribution of the debiased Lasso and show that a degrees-of-freedom correction is necessary for computing valid confidence intervals.

AoS Spatial & Time Series · 发表于 2023-10-01

Spatial quantiles on the hypersphere

Dimitri Konen, Davy Paindaveine

摘要

We propose a concept of quantiles for probability measures on the unit hypersphere Sd−1 of Rd. The innermost quantile is the Fréchet median, that is, the L1-analog of the Fréchet mean. The proposed quantiles μα,um are directional in nature: they are indexed by a scalar order α∈[0,1] and a unit vector u in the tangent space TmSd−1 to Sd−1 at m. To ensure computability in any dimension d, our quantiles are essentially obtained by considering the Euclidean (Chaudhuri (J. Amer. Statist. Assoc. 91 (1996) 862–872)) spatial quantiles in a suitable stereographic projection of Sd−1 onto TmSd−1. Despite this link with Euclidean spatial quantiles, studying the proposed spherical quantiles requires understanding the nature of the (Chaudhuri (1996)) quantiles in a version of the projective space where all points at infinity are identified. We thoroughly investigate the structural properties of our quantiles and we further study the asymptotic behavior of their sample versions, which requires controlling the impact of estimating m. Our spherical quantile concept also allows for companion concepts of ranks and depth on the hypersphere. We illustrate the relevance of our construction by considering two inferential applications, related to supervised classification and to testing for rotational symmetry.

AoS High-Dimensional · 发表于 2023-10-01

Sharp optimality for high-dimensional covariance testing under sparse signals

Song Xi Chen, Yumou Qiu, Shuyi Zhang

摘要

This paper considers one-sample testing of a high-dimensional covariance matrix by deriving the detection boundary as a function of the signal sparsity and signal strength under the sparse alternative hypotheses. It first shows that the optimal detection boundary for testing sparse means is the minimax detection lower boundary for testing the covariance matrix. A multilevel thresholding test is proposed and is shown to be able to attain the detection lower boundary over a substantial range of the sparsity parameter, implying that the multilevel thresholding test is sharp optimal in the minimax sense over the range. The asymptotic distribution of the multilevel thresholding statistic for covariance matrices is derived under both Gaussian and non-Gaussian distributions by developing a novel U-statistic decomposition in conjunction with the matrix blocking and the coupling techniques to handle the complex dependence among the elements of the sample covariance matrix. The superiority in the detection boundary of the multilevel thresholding test over the existing tests is also demonstrated.

AoS Testing & Uncertainty High-Dimensional · 发表于 2023-10-01

Robust high-dimensional tuning free multiple testing

Jianqing Fan, Zhipeng Lou, Mengxin Yu

摘要

A stylized feature of high-dimensional data is that many variables have heavy tails, and robust statistical inference is critical for valid large-scale statistical inference. Yet, the existing developments such as Winsorization, Huberization and median of means require the bounded second moments and involve variable-dependent tuning parameters, which hamper their fidelity in applications to large-scale problems. To liberate these constraints, this paper revisits the celebrated Hodges–Lehmann (HL) estimator for estimating location parameters in both the one- and two-sample problems, from a nonasymptotic perspective. Our study develops Berry–Esseen inequality and Cramér-type moderate deviation for the HL estimator based on newly developed nonasymptotic Bahadur representation and builds data-driven confidence intervals via a weighted bootstrap approach. These results allow us to extend the HL estimator to large-scale studies and propose tuning-free and moment-free high-dimensional inference procedures for testing global null and for large-scale multiple testing with false discovery proportion control. It is convincingly shown that the resulting tuning-free and moment-free methods control false discovery proportion at a prescribed level. The simulation studies lend further support to our developed theory.

AoS Nonparametric & Semiparametric · 发表于 2023-10-01

Optimal nonparametric testing of Missing Completely At Random and its connections to compatibility

Thomas B. Berrett, Richard J. Samworth

摘要

Given a set of incomplete observations, we study the nonparametric problem of testing whether data are Missing Completely At Random (MCAR). Our first contribution is to characterise precisely the set of alternatives that can be distinguished from the MCAR null hypothesis. This reveals interesting and novel links to the theory of Fréchet classes (in particular, compatible distributions) and linear programming, that allow us to propose MCAR tests that are consistent against all detectable alternatives. We define an incompatibility index as a natural measure of ease of detectability, establish its key properties and show how it can be computed exactly in some cases and bounded in others. Moreover, we prove that our tests can attain the minimax separation rate according to this measure, up to logarithmic factors. Our methodology does not require any complete cases to be effective, and is available in the R package MCARtest.

第 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 页