arXiv subjects

Maria-Pia Victoria-Feser

Publications and source records attributed to Maria-Pia Victoria-Feser.

14 recordsLinked to original sources

arXiv2024

Multivariate Adjustments for Average Equivalence Testing

Multivariate (average) equivalence testing is widely used to assess whether the means of two conditions of interest are `equivalent' for different outcomes simultaneously. The multivariate Two One-Sided Tests (TOST) procedure is typically used in this context by checking if, outcome by outcome, the marginal $100(1-2α$)\% confidence intervals for the difference in means between the two conditions of interest lie within pre-defined lower and upper equivalence limits. This procedure, known to be conservative in the univariate case, leads to a rapid power loss when the number of outcomes increases, especially when one or more outcome variances are relatively large. In this work, we propose a finite-sample adjustment for this procedure, the multivariate $α$-TOST, that consists in a correction of $α$, the significance level, taking the (arbitrary) dependence between the outcomes of interest into account and making it uniformly more powerful than the conventional multivariate TOST. We present an iterative algorithm allowing to efficiently define $α^{\star}$, the corrected significance level, a task that proves challenging in the multivariate setting due to the inter-relationship between $α^{\star}$ and the sets of values belonging to the null hypothesis space and defining the test size. We study the operating characteristics of the multivariate $α$-TOST both theoretically and via an extensive simulation study considering cases relevant for real-world analyses -- i.e.,~relatively small sample sizes, unknown and heterogeneous variances, and different correlation structures -- and show the superior finite-sample properties of the multivariate $α$-TOST compared to its conventional counterpart. We finally re-visit a case study on ticlopidine hydrochloride and compare both methods when simultaneously assessing bioequivalence for multiple pharmacokinetic parameters.

stat.ME↗

arXiv2024

An accurate percentile method for parametric inference based on asymptotically biased estimators

Inference methods for computing confidence intervals in parametric settings usually rely on consistent estimators of the parameter of interest. However, it may be computationally and/or analytically burdensome to obtain such estimators in various parametric settings, for example when the data exhibit certain features such as censoring, misclassification errors or outliers. To address these challenges, we propose a simulation-based inferential method, called the implicit bootstrap, that remains valid regardless of the potential asymptotic bias of the estimator on which the method is based. We demonstrate that this method allows for the construction of asymptotically valid percentile confidence intervals of the parameter of interest. Additionally, we show that these confidence intervals can also achieve second-order accuracy. We also show that the method is exact in three instances where the standard bootstrap fails. Using simulation studies, we illustrate the coverage accuracy of the method in three examples where standard parametric bootstrap procedures are computationally intensive and less accurate in finite samples.

stat.ME↗

arXiv2023

Just Identified Indirect Inference Estimator: Accurate Inference through Bias Correction

An important challenge in statistical analysis lies in controlling the estimation bias when handling the ever-increasing data size and model complexity of modern data settings. In this paper, we propose a reliable estimation and inference approach for parametric models based on the Just Identified iNdirect Inference estimator (JINI). The key advantage of our approach is that it allows to construct a consistent estimator in a simple manner, while providing strong bias correction guarantees that lead to accurate inference. Our approach is particularly useful for complex parametric models, as it allows to bypass the analytical and computational difficulties (e.g., due to intractable estimating equation) typically encountered in standard procedures. The properties of JINI (including consistency, asymptotic normality, and its bias correction property) are also studied when the parameter dimension is allowed to diverge, which provide the theoretical foundation to explain the advantageous performance of JINI in increasing dimensional covariates settings. Our simulations and an alcohol consumption data analysis highlight the practical usefulness and excellent performance of JINI when data present features (e.g., misclassification, rounding) as well as in robust estimation.

stat.ME↗

arXiv2020

Prevalence Estimation from Random Samples and Census Data with Participation Bias

Countries officially record the number of COVID-19 cases based on medical tests of a subset of the population with unknown participation bias. For prevalence estimation, the official information is typically discarded and, instead, small random survey samples are taken. We derive (maximum likelihood and method of moment) prevalence estimators, based on a survey sample, that additionally utilize the official information, and that are substantially more accurate than the simple sample proportion of positive cases. Put differently, using our estimators, the same level of precision can be obtained with substantially smaller survey samples. We take into account the possibility of measurement errors due to the sensitivity and specificity of the medical testing procedure. The proposed estimators and associated confidence intervals are implemented in the companion open source R package cape.

stat.ME↗

arXiv2020

A General Approach for Simulation-based Bias Correction in High Dimensional Settings

An important challenge in statistical analysis lies in controlling the bias of estimators due to the ever-increasing data size and model complexity. Approximate numerical methods and data features like censoring and misclassification often result in analytical and/or computational challenges when implementing standard estimators. As a consequence, consistent estimators may be difficult to obtain, especially in complex and/or high dimensional settings. In this paper, we study the properties of a general simulation-based estimation framework that allows to construct bias corrected consistent estimators. We show that the considered approach leads, under more general conditions, to stronger bias correction properties compared to alternative methods. Besides its bias correction advantages, the considered method can be used as a simple strategy to construct consistent estimators in settings where alternative methods may be challenging to apply. Moreover, the considered framework can be easily implemented and is computationally efficient. These theoretical results are highlighted with simulation studies of various commonly used models, including the negative binomial regression (with and without censoring) and the logistic regression (with and without misclassification errors). Additional numerical illustrations are provided in the supplementary materials.

math.ST↗

arXiv2020

Asymptotically Optimal Bias Reduction for Parametric Models

An important challenge in statistical analysis concerns the control of the finite sample bias of estimators. This problem is magnified in high-dimensional settings where the number of variables $p$ diverges with the sample size $n$, as well as for nonlinear models and/or models with discrete data. For these complex settings, we propose to use a general simulation-based approach and show that the resulting estimator has a bias of order $\mathcal{O}(0)$, hence providing an asymptotically optimal bias reduction. It is based on an initial estimator that can be slightly asymptotically biased, making the approach very generally applicable. This is particularly relevant when classical estimators, such as the maximum likelihood estimator, can only be (numerically) approximated. We show that the iterative bootstrap of Kuk (1995) provides a computationally efficient approach to compute this bias reduced estimator. We illustrate our theoretical results in simulation studies for which we develop new bias reduced estimators for the logistic regression, with and without random effects. These estimators enjoy additional properties such as robustness to data contamination and to the problem of separability.

math.ST↗

arXiv2020

Robust Two-Step Wavelet-Based Inference for Time Series Models

Complex time series models such as (the sum of) ARMA$(p,q)$ models with additional noise, random walks, rounding errors and/or drifts are increasingly used for data analysis in fields such as biology, ecology, engineering and economics where the length of the observed signals can be extremely large. Performing inference on and/or prediction from these models can be highly challenging for several reasons: (i) the data may contain outliers that can adversely affect the estimation procedure; (ii) the computational complexity can become prohibitive when models include more than just a few parameters and/or the time series are large; (iii) model building and/or selection adds another layer of (computational) complexity to the previous task; and (iv) solutions that address (i), (ii) and (iii) simultaneously do not exist in practice. For this reason, this paper aims at jointly addressing these challenges by proposing a general framework for robust two-step estimation based on a bounded influence M-estimator of the wavelet variance. In this perspective, we first develop the conditions for the joint asymptotic normality of the latter estimator thereby providing the necessary tools to perform (direct) inference for scale-based analysis of signals. Taking advantage of the model-independent weights of this first-step estimator that are computed only once, we then develop the asymptotic properties of two-step robust estimators using the framework of the Generalized Method of Wavelet Moments (GMWM), hence defining the Robust GMWM (RGMWM) that we then use for robust model estimation and inference in a computationally efficient manner even for large time series. Simulation studies illustrate the good finite sample performance of the RGMWM estimator and applied examples highlight the practical relevance of the proposed approach.

stat.ME↗

arXiv2019

Phase Transition Unbiased Estimation in High Dimensional Settings

An important challenge in statistical analysis concerns the control of the finite sample bias of estimators. For example, the maximum likelihood estimator has a bias that can result in a significant inferential loss. This problem is typically magnified in high-dimensional settings where the number of variables $p$ is allowed to diverge with the sample size $n$. However, it is generally difficult to establish whether an estimator is unbiased and therefore its asymptotic order is a common approach used (in low-dimensional settings) to quantify the magnitude of the bias. As an alternative, we introduce a new and stronger property, possibly for high-dimensional settings, called phase transition unbiasedness. An estimator satisfying this property is unbiased for all $n$ greater than a finite sample size $n^\ast$. Moreover, we propose a phase transition unbiased estimator built upon the idea of matching an initial estimator computed on the sample and on simulated data. It is not required for this initial estimator to be consistent and thus it can be chosen for its computational efficiency and/or for other desirable properties such as robustness. This estimator can be computed using a suitable simulation based algorithm, namely the iterative bootstrap, which is shown to converge exponentially fast. In addition, we demonstrate the consistency and the limiting distribution of this estimator in high-dimensional settings. Finally, as an illustration, we use our approach to develop new estimators for the logistic regression model, with and without random effects, that also enjoy other properties such as robustness to data contamination and are also not affected by the problem of separability. In a simulation exercise, the theoretical results are confirmed in settings where the sample size is relatively small compared to the model dimension.

math.ST↗

arXiv2019

A simple recipe for making accurate parametric inference in finite sample

Constructing tests or confidence regions that control over the error rates in the long-run is probably one of the most important problem in statistics. Yet, the theoretical justification for most methods in statistics is asymptotic. The bootstrap for example, despite its simplicity and its widespread usage, is an asymptotic method. There are in general no claim about the exactness of inferential procedures in finite sample. In this paper, we propose an alternative to the parametric bootstrap. We setup general conditions to demonstrate theoretically that accurate inference can be claimed in finite sample.

stat.ME↗

arXiv2018

On the Properties of Simulation-based Estimators in High Dimensions

Considering the increasing size of available data, the need for statistical methods that control the finite sample bias is growing. This is mainly due to the frequent settings where the number of variables is large and allowed to increase with the sample size bringing standard inferential procedures to incur significant loss in terms of performance. Moreover, the complexity of statistical models is also increasing thereby entailing important computational challenges in constructing new estimators or in implementing classical ones. A trade-off between numerical complexity and statistical properties is often accepted. However, numerically efficient estimators that are altogether unbiased, consistent and asymptotically normal in high dimensional problems would generally be ideal. In this paper, we set a general framework from which such estimators can easily be derived for wide classes of models. This framework is based on the concepts that underlie simulation-based estimation methods such as indirect inference. The approach allows various extensions compared to previous results as it is adapted to possibly inconsistent estimators and is applicable to discrete models and/or models with a large number of parameters. We consider an algorithm, namely the Iterative Bootstrap (IB), to efficiently compute simulation-based estimators by showing its convergence properties. Within this framework we also prove the properties of simulation-based estimators, more specifically the unbiasedness, consistency and asymptotic normality when the number of parameters is allowed to increase with the sample size. Therefore, an important implication of the proposed approach is that it allows to obtain unbiased estimators in finite samples. Finally, we study this approach when applied to three common models, namely logistic regression, negative binomial regression and lasso regression.

math.ST↗

arXiv2015

A Prediction Divergence Criterion for Model Selection

The problem of model selection is inevitable in an increasingly large number of applications involving partial theoretical knowledge and vast amounts of information, like in medicine, biology or economics. The associated techniques are intended to determine which variables are "important" to "explain a phenomenon under investigation. The terms "important" and "explain" can have very different meanings according to the context and, in fact, model selection can be applied to any situation where one tries to balance variability with complexity. In this paper, we introduce a new class of error measures and of model selection criteria, to which many well know selection criteria belong. Moreover, this class enables us to derive a novel criterion, based on a divergence measure between the predictions produced by two nested models, called the Prediction Divergence Criterion (PDC). Our selection procedure is developed for linear regression models, but has the potential to be extended to other models. We demonstrate that, under some regularity conditions, it is asymptotically loss efficient and can also be consistent. In the linear case, the PDC is a counterpart to Mallow's Cp but with a lower asymptotic probability of overfitting. In a case study and by means of simulations, the PDC is shown to be particularly well suited in "sparse" settings with correlated covariates which we believe to be common in real applications.

stat.ME↗

arXiv2013

Robust VIF regression with application to variable selection in large data sets

The sophisticated and automated means of data collection used by an increasing number of institutions and companies leads to extremely large data sets. Subset selection in regression is essential when a huge number of covariates can potentially explain a response variable of interest. The recent statistical literature has seen an emergence of new selection methods that provide some type of compromise between implementation (computational speed) and statistical optimality (e.g., prediction error minimization). Global methods such as Mallows' $C_p$ have been supplanted by sequential methods such as stepwise regression. More recently, streamwise regression, faster than the former, has emerged. A recently proposed streamwise regression approach based on the variance inflation factor (VIF) is promising, but its least-squares based implementation makes it susceptible to the outliers inevitable in such large data sets. This lack of robustness can lead to poor and suboptimal feature selection. In our case, we seek to predict an individual's educational attainment using economic and demographic variables. We show how classical VIF performs this task poorly and a robust procedure is necessary for policy makers. This article proposes a robust VIF regression, based on fast robust estimators, that inherits all the good properties of classical VIF in the absence of outliers, but also continues to perform well in their presence where the classical approach fails.

stat.AP↗

arXiv2011

Zero-inflated truncated generalized Pareto distribution for the analysis of radio audience data

Extreme value data with a high clump-at-zero occur in many domains. Moreover, it might happen that the observed data are either truncated below a given threshold and/or might not be reliable enough below that threshold because of the recording devices. These situations occur, in particular, with radio audience data measured using personal meters that record environmental noise every minute, that is then matched to one of the several radio programs. There are therefore genuine zeros for respondents not listening to the radio, but also zeros corresponding to real listeners for whom the match between the recorded noise and the radio program could not be achieved. Since radio audiences are important for radio broadcasters in order, for example, to determine advertisement price policies, possibly according to the type of audience at different time points, it is essential to be able to explain not only the probability of listening to a radio but also the average time spent listening to the radio by means of the characteristics of the listeners. In this paper we propose a generalized linear model for zero-inflated truncated Pareto distribution (ZITPo) that we use to fit audience radio data. Because it is based on the generalized Pareto distribution, the ZITPo model has nice properties such as model invariance to the choice of the threshold and from which a natural residual measure can be derived to assess the model fit to the data. From a general formulation of the most popular models for zero-inflated data, we derive our model by considering successively the truncated case, the generalized Pareto distribution and then the inclusion of covariates to explain the nonzero proportion of listeners and their average listening time. By means of simulations, we study the performance of the maximum likelihood estimator (and derived inference) and use the model to fully analyze the audience data of a radio station in a certain area of Switzerland.

stat.AP↗

arXiv2009

Assessing multivariate predictors of financial market movements: A latent factor framework for ordinal data

Much of the trading activity in Equity markets is directed to brokerage houses. In exchange they provide so-called "soft dollars," which basically are amounts spent in "research" for identifying profitable trading opportunities. Soft dollars represent about USD 1 out of every USD 10 paid in commissions. Obviously they are costly, and it is interesting for an institutional investor to determine whether soft dollar inputs are worth being used (and indirectly paid for) or not, from a statistical point of view. To address this question, we develop association measures between what broker--dealers predict and what markets realize. Our data are ordinal predictions by two broker--dealers and realized values on several markets, on the same ordinal scale. We develop a structural equation model with latent variables in an ordinal setting which allows us to test broker--dealer predictive ability of financial market movements. We use a multivariate logit model in a latent factor framework, develop a tractable estimator based on a Laplace approximation, and show its consistency and asymptotic normality. Monte Carlo experiments reveal that both the estimation method and the testing procedure perform well in small samples. The method is then used to analyze our dataset.

stat.AP↗