arXiv subjects

Rahul Ghosal

Publications and source records attributed to Rahul Ghosal.

24 records · Page 2Linked to original sources

arXiv2021

Bayesian Inference for Generalized Linear Model with Linear Inequality Constraints

Bayesian statistical inference for Generalized Linear Models (GLMs) with parameters lying on a constrained space is of general interest (e.g., in monotonic or convex regression), but often constructing valid prior distributions supported on a subspace spanned by a set of linear inequality constraints can be challenging, especially when some of the constraints might be binding leading to a lower dimensional subspace. For the general case with canonical link, it is shown that a generalized truncated multivariate normal supported on a desired subspace can be used. Moreover, it is shown that such prior distribution facilitates the construction of a general purpose product slice sampling method to obtain (approximate) samples from corresponding posterior distribution, making the inferential method computationally efficient for a wide class of GLMs with an arbitrary set of linear inequality constraints. The proposed product slice sampler is shown to be uniformly ergodic, having a geometric convergence rate under a set of mild regularity conditions satisfied by many popular GLMs (e.g., logistic and Poisson regressions with constrained coefficients). One of the primary advantages of the proposed Bayesian estimation method over classical methods is that uncertainty of parameter estimates is easily quantified by using the samples simulated from the path of the Markov Chain of the slice sampler. Numerical illustrations using simulated data sets are presented to illustrate the superiority of the proposed methods compared to some existing methods in terms of sampling bias and variances. In addition, real case studies are presented using data sets for fertilizer-crop production and estimating the SCRAM rate in nuclear power plants.

stat.ME↗

arXiv2021

Scalar on time-by-distribution regression and its application for modelling associations between daily-living physical activity and cognitive functions in Alzheimer's Disease

Wearable data is a rich source of information that can provide deeper understanding of links between human behaviours and human health. Existing modelling approaches use wearable data summarized at subject level via scalar summaries using regression techniques, temporal (time-of-day) curves using functional data analysis (FDA), and distributions using distributional data analysis (DDA). We propose to capture temporally local distributional information in wearable data using subject-specific time-by-distribution (TD) data objects. Specifically, we propose scalar on time-by-distribution regression (SOTDR) to model associations between scalar response of interest such as health outcomes or disease status and TD predictors. We show that TD data objects can be parsimoniously represented via a collection of time-varying L-moments that capture distributional changes over the time-of-day. The proposed method is applied to the accelerometry study of mild Alzheimer's disease (AD). Mild AD is found to be significantly associated with reduced maximal level of physical activity, particularly during morning hours. It is also demonstrated that TD predictors attain much stronger associations with clinical cognitive scales of attention, verbal memory, and executive function when compared to predictors summarized via scalar total activity counts, temporal functional curves, and quantile functions. Taken together, the present results suggest that the SOTDR analysis provides novel insights into cognitive function and AD.

stat.AP↗

arXiv2019

A Score Based Test for Functional Linear Concurrent Regression

We propose a novel method for testing the null hypothesis of no effect of a covariate on the response in the context of functional linear concurrent regression. We establish an equivalent random effects formulation of our functional regression model under which our testing problem reduces to testing for zero variance component for random effects. For this purpose, we use a one-sided score test approach, which is an extension of the classical score test. We provide theoretical justification as to why our testing procedure has the right levels (asymptotically) under null using standard assumptions. Using numerical simulations, we show that our testing method has the desired type I error rate and gives higher power compared to a bootstrapped F test currently existing in the literature. Our model and testing procedure are shown to give good performances even when the data is sparsely observed, and the covariate is contaminated with noise. Applications of the proposed testing method are demonstrated on gait study and a dietary calcium absorption data.

stat.ME↗

arXiv2019

Variable Selection in Functional Linear Concurrent Regression

We propose a novel method for variable selection in functional linear concurrent regression. Our research is motivated by a fisheries footprint study where the goal is to identify important time-varying socio-structural drivers influencing patterns of seafood consumption, and hence fisheries footprint, over time, as well as estimating their dynamic effects. We develop a variable selection method in functional linear concurrent regression extending the classically used scalar on scalar variable selection methods like LASSO, SCAD, and MCP. We show in functional linear concurrent regression the variable selection problem can be addressed as a group LASSO, and their natural extension; group SCAD or a group MCP problem. Through simulations, we illustrate our method, particularly with group SCAD or group MCP penalty, can pick out the relevant variables with high accuracy and has minuscule false positive and false negative rate even when data is observed sparsely, is contaminated with noise and the error process is highly non-stationary. We also demonstrate two real data applications of our method in studies of dietary calcium absorption and fisheries footprint in the selection of influential time-varying covariates.

stat.AP↗

arXiv2019

Estimating menarcheal age distribution from partially recalled data

In a cross-sectional study, adolescent and young adult females were asked to recall the time of menarche, if experienced. Some respondents recalled the date exactly, some recalled only the month or the year of the event, and some were unable to recall anything. We consider estimation of the menarcheal age distribution from this interval censored data. A~complicated interplay between age-at-event and calendar time, together with the evident fact of memory fading with time, makes the censoring informative. We propose a model where the probabilities of various types of recall would depend on the time since menarche. For parametric estimation we model these probabilities using multinomial regression function. Establishing consistency and asymptotic normality of the parametric MLE requires a bit of tweaking of the standard asymptotic theory, as the data format varies from case to case. We also provide a non-parametric MLE, propose a computationally simpler approximation, and establish the consistency of both these estimators under mild conditions. We study the small sample performance of the parametric and non-parametric estimators through Monte Carlo simulations. Moreover, we provide a graphical check of the assumption of the multinomial model for the recall probabilities, which appears to hold for the menarcheal data set. Our analysis shows that the use of the partially recalled part of the data indeed leads to smaller confidence intervals of the survival function.

stat.ME↗

arXiv2018

A Statistical Exploration of Duckworth-Lewis Method Using Bayesian Inference

Duckworth-Lewis (D/L) method is the incumbent rain rule used to decide the result of a limited overs cricket match should it not be able to reach its natural conclusion. Duckworth and Lewis (1998) devised a two factor relationship between the numbers of overs a team had remaining and the number of wickets they had lost in order to quantify the percentage resources a team has at any stage of the match. As number of remaining overs decrease and lost wickets increase the resources are expected to decrease. The resource table which is still being used by ICC (International Cricket Council) for 50 overs cricket match suffers from lack of monotonicity both in numbers of overs left and number of wickets lost. We apply Bayesian inference to build a resource table which overcomes the non monotonicity problem of the current D/L resource table and show that it gives better prediction for teams in first innings score and hence it is more suitable for using in rain affected matches.

stat.AP↗