MAKE MOST OF THE KNOWLEDGE NETWORK, JOIN ACADEMIC RESEARCH FOUNDATION

The paper studies model selection through consistency of several Bayes factors for the Black-Karasinski models with proper informative prior and noninformative prior. We consider continuous and discrete (both dense and fixed time intervals) observations of the process.

KEYWORDS: Black-Karasinski model, Ornstein-Uhlenbeck process, Vasicek model, Bayes factor, mean reversion, model choice, consistency, proper prior, noninformative prior, continuous observation, discrete observations.

2020 Mathematics Subject Classification: 60F05, 60H10, 62F12, 62F15, 62M05, 91G30, 91G60

The paper deals with four robust partial correlation coefficients. Two methods for testing the hypothesis of a zero correlation are studied via simulations. One of these methods deals with contamination bias, which refers to the fact that many robust regression estimators that have a high breakdown can be negatively impacted by few bad leverage points. The methods considered are illustrated using data that deals with playfulness and empathy.

KEYWORDS: skipped correlation; Winsorized correlation; Spearman’s rho; Kendall’s tau; MM-estimator

We introduce a skewed version of the continuous uniform distribution based on a general procedure to extend distributions proposed by Cort´es et al. [Journal of Probability and Statistics, Vol. 2018, pp. 1-10.]. We determine the density and the cumulative distribution function of the new distribution and study diverse properties like moments, moment generating function, shape characteristics, quantile function, Hazard rate, entropy and maximum likelihood estimators. We apply the new model to describe COVID-19 mortality rates and online product reviews.

KEYWORDS: Skewed distributions; distribution theory; Hazard rate function; Shannon entropy; COVID-19 death rates; product reviews

Background: Based on an increased prevalence of diabetes, asthma and hypertension among women in reproductive age, understanding the risk factors of noncommunicable diseases (NCDs) is crucial to inform policy and program interventions to address the problem. Objectives: This study is to examine the prevalence and determinants of noncommunicable diseases across socioeconomic characteristics among reproductive age women. The data is derived from the National Family Health Survey conducted in 2019–2021. The effective sample size for the present paper was 72415 women aged 15-49 years in India.

Methods: Descriptive statistics along with bivariate analysis were conducted to find the preliminary results. Additionally, multi-variable logistic regression analysis is conducted to find the relationship between NCDs and behavioral factors such as alcohol consumption and tobacco use in the present study.

Results: It is revealed about 10% of women had any of the NCDs followed by 4.7% hypertension, 2.7% thyroid, 1.9% diabetes, 1.6% Asthama, 0.7% heart diseases, 0.6% chronic kidney diseases and 0.1% Cancer. A proportion of 16.4% of women smoked tobacco whereas 12.8% of women consumed chew tobacco. Also, a proportion of 10.3% of women consumed alcohol in the study. The odds of having NCDs among women who chew tobacco and consume alcohol are 1.40 times higher odds of having NCDs among women compared to its counterparts.

Conclusion: Screening and early detection of the same non-communicable diseases such as hypertension and diabetes are especially emphasized among people who smoke and chew tobacco and consume alcohol. In addition, interventions focused on correctable factors such as smoking drinking and the associated obesity may help prevent an increased burden on NCD in Indian women.

Time series models with non-Gaussian marginal distributions have received much attention in recent years. This include autoregressive models with exponential, Pareto, Weibull, logistic, mixed Gamma, Laplace, Cauchy and Stable marginal distributions. In this paper, autoregressive models with Type I and Type II generalized Geometric Linnik marginal are developed.

KEYWORDS: Geometric Linnik Distribution, Generalized Geometric Linnik Distribution, Autoregressive Models