Brazilian Journal of Biometrics

Assessment of the evolution of patients hospitalized for COVID-19 in Paraná

José Bruno Pereira Souza — 2025-02-11

The COVID-19 pandemic was marked by great fear, as it was a new disease of which we had no knowledge of its effects and prevention methods. However, during this period,we also made significant advances in research across various fields, from studying the causes and effects of the disease to the development of vaccines. In this study, we focus on assessing the evolution (recovery/death) of COVID-19 inpatients, who were hospitalized in the state of Paraná, Brazil in the year of 2022. To achieve this, we analyzed data from the System of Epidemiological Surveillance of Influenza (SIVEP) which provides information about Brazilian patients hospitalized with severe acute respiratory syndrome, using several machine learning techniques that allowed us to relate the patient evolution to possible associated factors. Results showed that age, gender, education, and neurological disorder, among other factors, have significant impacts in the inpatients evolution. When predicting the patient outcome, we obtained an accuracy over 75%, which shows the efficiency of the models.

A novel ratio cum product type exponential class of estimators of finite population mean in Adaptive cluster Sampling

Rohan Mishra — 2025-02-11

In the present paper a ratio cum product type exponential class of estimators has been proposed to estimate the finite population mean of rare type or hard to reach type population. The mean square error and bias expressions of the proposed generalized class have been derived and presented up to the first order of approximation. New estimators have been developed from the proposed class using robust measures. Using simulation study and a real data application, the efficiency of the newly developed estimators from the class that is proposed have been shown. The results show that the new developed estimators are more efficient than the competing estimators presented in this paper.

An application of the k-means clustering method to National Public Safety Data - the case of intentional homicides

Henrique José de Paula Alves — 2025-02-11

The purpose of this article is to assist Brazilian authorities in directing resources to reduce the rate of intentional homicide per 100,000 inhabitants in the federative units of Brazil. In this sense, linked to the information made available by the National Secretariat for Public Security (Senasp), it was decided to apply the grouping technique known as k-means to group the FU’s by similar rates. Three clusters were found and, then, a detailed description of each of the FU’s belonging to each of the three clusters was carried out. We noticed that only the FU’s that form cluster 1 have a homicide rate below 16 victims per 100,000 inhabitants, the limit imposed by the 2021 National Public Security Plan.. This plan was prepared by Brazilian Ministry of Justice, Senasp and state public security departments. The information provided by Senasp was accounted for in the period from 2015 to 2022 and the rates were calculated, in the usual way, for each year.

The Cause and Trend of Contraceptive Discontinuation in India: A Comprehensive Analysis Employing a Multiple Decrement Model

Abhay Kumar Tiwari — 2025-02-11

The usage of contraceptives and family planning are hot issues of conversation due to the growing population. Here authors have done a comprehensive study on contraceptive discontinuation in India. This article describes the trend in contraceptive discontinuation in India and the multiple decrement concept is used to determine the probability of contraceptive discontinuation in India. Data used in this research were taken from the country’s National Family and Health Survey. The contraceptive discontinuation trend reveals that from NFHS 1 to NFHS 5 there is a decline of 15.25% in contraceptive discontinuation in India. It was realized that in NFHS 1, the highest discontinuation was due to fertility-related causes, and in NFHS 5, it was due to method-related causes. The age groups where the highest discontinuation occurred were 40-49 and 25-29 in NFHS 1 and NFHS 5, respectively.

Residual analysis for discrete correlated data in the multivariate approach

Lizandra C. Fábio — 2025-02-11

The residual distributions obtained from discrete correlated and uncorrelated data cannot be well approximated to the standardized normal distribution. In this case, the efficiency in checking the adequacy of the model to the data and detecting outliers is not guaranteed. Thus, alternative measures for residual analysis have been considered in several classes of models and their properties have been assessed. In this paper, we investigate the empirical distribution of four residuals of the multivariate negative binomial regression (MNBR) model. In our study, we propose standardized weighted and standardized Pearson residuals; we also consider the standardized component of deviance and quantile residuals suggested by Fabio et al. (2012) and Fabio et al. (2023), respectively. Monte Carlo simulation results reveal that the concordance of the empirical distribution of the residuals to the standard normal distribution depends on the dispersion parameter. Furthermore, the impact on residual analysis when the random effect distribution is misspecified is explored. We concluded that the quantile and standardized weighted residuals presented better performances.

Procrustes analysis, multivariate regression, variable selection and outlier detection in compositional data for social vulnerability

Paulo Meira e Silva de Oliveira — 2025-02-11

Vulnerability means delicate and weak in the behavior of people, objects, situations and ideas. People considered “socially vulnerable” are those who lose their representation in society and generally depend on help from third parties to ensure survival. The main characteristics that mark this vulnerability are precarious housing conditions, sanitation, non-existent means of subsistence and the absence of a family environment. Among the different types, they highlight youth in the area of health, marginalization, exclusion and territorial. Social Vulnerability Index (SVI) is composed of indicators of income and social impairment in dimensions such as identification, housing, education, income, poverty, family, work and other assets. Variable selection is finding a subset of variables that best explains a response vector, without losing relevant information. Procrustes Analysis is a method that aims to determine how much a subset of variables best represents the structure of the original data. Compositional data are quantitative descriptions of the parts of a whole, which convey information in a relative way. Principal components are linear combinations of all original variables, independent of each other and estimated with the purpose of retaining, in order of estimation, the maximum amount of information to explain the total variance. Univariate outliers are observations that differ greatly from the others. Multivariate outlier corresponds to cases involving two or more variables. In this work we use the Procrustes method and other regression methods to select variables formed from compositional data after detecting multivariate outliers using Mahalanobis Distance and comedian approach.

A general approach to the isobolographic method

Vladimir Panov — 2025-02-11

Strict definitions and formal mathematical constructions are given to represent the main concepts of the isobolographic method as mathematical objects. In particular, a strict definition of zero interaction notion is introduced. The peculiarity of this definition is that this notion appears to depend on the dose-response function of a particular acting agent, whereas it is commonly believed that it is completely determined only by the whole set of acting agents. It is shown that without additional assumptions about the type of dose-response functions, a type of joint action of agents can be different and even opposite depending on the dose-response function of which the notion of zero interaction is considered. The only case when the concept of zero interaction is unambiguously defined and does not depend on the chosen dose-response
function is the case of scale equivalence of dose-response functions of all acting agents. A theorem on the representation of the zero-interaction manifold in the case of arbitrary single-factor dose-response functions is proved. Examples of analyzing the joint action of factors using isoboles for a two-factor linear model with a cross term and a quadratic model are considered.

A New Model of Mixture Distribution Using a Survival Analysis of Cancer Patients

M. Sakthivel — 2025-02-11

In this article, specific statistical considerations are typically required in order to select the best model for fitting cancer survival data. A new two-parameter distribution known as the Mixture of Lomax and Gamma Distribution (MLGD) is proposed in this article. Because of the unique way that the Gamma and Lomax distributions are mixed, this distribution is created as a special mixture of two distributions. Statistical properties, order statistics, entropy, and reliability analysis are also derived. The maximum likelihood estimation method can be used to estimate the parameters of the distribution. Lastly, a goodness-of-fit analysis is demonstrated on a set of data on cancer survival. It is compared to the fit and shows that the gamma and Lomax mixing distributions have more flexibility than the other distributions.