Vandna Jowaheer

Estimating stature from percutaneous length of tibia and ulna in Indo-Mauritian population

Stature estimation from percutaneous body measurements forms part of forensic anthropological analysis for the purpose of identification. This study is aimed at modeling the stature on the basis of percutaneous tibial and/or ulnar length in human subjects comprising of Indo-Mauritian population. The study was conducted in the Department of Forensic Medicine and Toxicology, SSR Medical College, Mauritius on 180 young and healthy students comprising of 90 males and 90 females in the age group ranging from 18 to 28 years. The measurements were taken by using standard anthropometric instruments. It is remarked that tibial as well as ulnar length show a linear relationship with the stature, where stature is normally distributed. However, it is required to transform the measurements on stature to obtain appropriate regression equations. Moreover, since ulna and tibia are significantly correlated, it is recommended to use the sum of the ulnar and tibial length to estimate stature, in case both are available for an individual. Our regression models are sufficiently validated and highly efficient.

Sex identification on the basis of hand and foot measurements in Indo-Mauritian population – A model based approach

Identification is the foremost issue in crime investigation. A few studies have been performed so far in order to identify sex on the basis of single foot or hand of the victim. Moreover, these studies provide only crude measures to indicate sex and there exists no concrete methodology to predict sex using the available information. In the present paper, we have developed statistical models to identify sex based on the dimensions of foot and hand. The models containing both length and breadth of hand or foot as independent variables are capable of predicting sex in Indo-Mauritian population with fairly high accuracy as compared to those containing hand or foot indices.

Modelling a non-stationary BINAR(1) Poisson process

ABSTRACT Non-stationarity in bivariate time series of counts may be induced by a number of time-varying covariates affecting the bivariate responses due to which the innovation terms of the individual series as well as the bivariate dependence structure becomes non-stationary. So far, in the existing models, the innovation terms of individual INAR(1) series and the dependence structure are assumed to be constant even though the individual time series are non-stationary. Under this assumption, the reliability of the regression and correlation estimates is questionable. Besides, the existing estimation methodologies such as the conditional maximum likelihood (CMLE) and the composite likelihood estimation are computationally intensive. To address these issues, this paper proposes a BINAR(1) model where the innovation series follow a bivariate Poisson distribution under some non-stationary distributional assumptions. The method of generalized quasi-likelihood (GQL) is used to estimate the regression effects while the serial and bivariate correlations are estimated using a robust moment estimation technique. The application of model and estimation method is made in the simulated data. The GQL method is also compared with the CMLE, generalized method of moments (GMM) and generalized estimating equation (GEE) approaches where through simulation studies, it is shown that GQL yields more efficient estimates than GMM and equally or slightly more efficient estimates than CMLE and GEE.

Comparing Joint GQL Estimation and GMM Adaptive Estimation in COM-Poisson Longitudinal Regression Model

It is of scientific interest to study the application of COM-Poisson model to the case of longitudinal response data, the analysis of which is quite challenging due to the fact that longitudinal responses of a subject are correlated and the correlation pattern is usually unknown. In this article, we extend the COM-Poisson GLM to the generalized linear longitudinal model. We also develop a joint generalized quasi-likelihood estimating equation approach based on a stationary autocorrelation structure for the repeated count data. We further compare the performance of this estimation method with that of Generalized Method of Moments through a simulation study.

An analysis of fingerprint ridge density in the Indo-Mauritian population and its application to gender determination:

Gender determination is an important aspect of personal identification, which is often required in medicolegal practice. Many experts believe that there are finer and more epidermal ridges on the fingers of women as compared with men. However, it is important to establish numerical cut-off values in terms of ridge counts to facilitate the gender determination within a particular population. The present study was conducted in the Department of Forensic Medicine, SSR Medical College, Mauritius with the objective to describe the ridge density in the Indo-Mauritian population and to devise a numerical model which is capable of identifying the sex of an individual from this population on the basis of the ridge counts obtained from the corresponding finger prints. The study was focused on 200 healthy medical students (100 men and 100 women) within the age range of 20–30 years. Multivariate analysis of variance results shows a significant gender difference in the sense that women tend to have higher ridge density then men in the distal region of all 10 digits (F = 41.83, P ≤ 0.005). The maximum mean ridge density over all fingers in men (12.26∼12) is less than the minimum mean ridge density over all fingers in women (12.71∼13). A linear discriminant function is derived from numerical modelling, which is used to predict (with a high reliability index, 0.92) the sex of the person whose fingerprints are obtained.

On familial longitudinal Poisson mixed models with gamma random effects

Poisson mixed models are used to analyze a wide variety of cluster count data. These models are commonly developed based on the assumption that the random effects have either the log-normal or the gamma distribution. Obtaining consistent as well as efficient estimates for the parameters involved in such Poisson mixed models has, however, proven to be difficult. Further problem gets mounted when the data are collected repeatedly from the individuals of the same cluster or family. In this paper, we introduce a generalized quasilikelihood approach to analyze the repeated familial data based on the familial structure caused by gamma random effects. This approach provides estimates of the regression parameters and the variance component of the random effects after taking the longitudinal correlations of the data into account. The estimators are consistent as well as highly efficient.

Estimating the parameters of a BINMA Poisson model for a non-stationary bivariate time series

ABSTRACT This article proposes a novel non-stationary BINMA time series model by extending two INMA processes where their innovation series follow the bivariate Poisson under time-varying moment assumptions. This article also demonstrates, through simulation studies, the use and superiority of the generalized quasi-likelihood (GQL) approach to estimate the regression effects, which is computationally less complicated as compared to conditional maximum likelihood estimation (CMLE) and the feasible generalized least squares (FGLS). The serial and bivariate dependence correlations are estimated by a robust method of moments.

A GQL estimation approach for analysing non-stationary over-dispersed BINAR(1) time series

ABSTRACT This paper proposes a generalized quasi-likelihood (GQL) function for estimating the vector of regression and over-dispersion effects for the respective series in the bivariate integer-valued autoregressive process of order 1 (BINAR(1)) with Negative Binomial (NB) marginals. The auto-covariance function in the proposed GQL is computed using some ‘robust’ working structures. As for the BINAR(1) process, the inter-relation between the series is induced mainly by the correlated NB innovations that are subject to different levels of over-dispersion. The performance of the GQL approach is tested via some Monte-Carlo simulations under different combination of over-dispersion together with low and high serial- and cross-correlation parameters. The model is also applied to analyse a real-life series of day and night accidents in Mauritius.

Analyzing longitudinal count data from adaptive clinical trials: a weighted generalized quasi-likelihood approach

In an adaptive clinical trial set up, there exist some adaptive designs to assign an incoming individual to a treatment so that more study subjects are assigned to the better treatment. These designs are however developed under the assumption that an individual patient provides a single response. In practice, there are situations where an individual assigned to a treatment may be required to provide a repeated number of responses over a period of time. Recently, Sutradhar et al. [Sutradhar, B.C., Biswas, A., and Bari, W., 2005, Marginal regression for binary longitudinal data in adaptive clinical trials. Scandinavian Journal of Statistics, 32, 93–113.] have proposed a simple longitudinal play-the-winner (SLPW) design as a generalization of the existing simple play-the-winner (SPW) design, in order to assign an incoming individual to a better treatment, under the binary longitudinal set up. In this paper, we deal with the longitudinal count responses and examine, through a simulation study, the performances of the SLPW design and a new bivariate random walk type design in allocating an individual patient to the better treatment group. As far as the estimation of the parameters is concerned, we examine the performance of a weighted generalized quasi-likelihood approach in estimating the parameters of the longitudinal model including the treatment effects.

Inferential methods for an unconstrained nonstationary BINMA time series process with Poisson innovations

This article proposes an unconstrained nonstationary BINMA(l) time-series process with Poisson innovations under time-dependent moments where the cross-correlation structure is formed firstly by the jointly distributed innovations and second by relating the current varíate observations with the previous lagged innovation of the other series and vice versa. For this new BINMA(1) time series model, feasible generalized least squares (FGLS), generalized method of moments (GMM), and generalized quasi-likelihood (GQL) estimating equations are developed. A simulation process is set up to generate BINMA(l) time-series data under the unconstrained cross-correlation structure. The purpose here is to assess the performance of the different estimation techniques proposed. The article also analyzes real-life monthly day and night accidents data in Mauritius under this model.