This paper investigates the inverse and power Juchez distribution. Some relevant properties are studied, and the distributions can altogether model many varieties of datasets emanating from different life phenomena. And then, a generator for other generalized distributions termed Juchez-G is developed to suffice for the relevance of robust model development. Finally, IJD and PJD showed to be a better fit over both the baseline distribution and their respective counterpart distributions, with respect to the datasets used. The proposed method is interesting and the performance is superior.

A novel distribution called the Juchez distribution is proposed and studied. This distribution is composite of both exponential and gamma distributions. The properties and features of this distribution are studied, with empirical emphasis: on the inequality relationship within the measures of central tendency, and the coefficient of variation. The model parameter was estimated using the method of maximum likelihood, where the asymptotic and consistent properties are numerically studied as well. The flexibility of this distribution is shown, through an application to a facebook Live-Streaming and Cancer data set. This distribution shows a high efficiency when compared with other one parameter distributions.

This is a comparative study on mixture distribution; where the study seeks to ascertain whether higher number of k-component mixtures could result to development of models that show better fits. In the performance comparison, special consideration was given to univariate one parameter distributions derived using mixture models, and the results show that distributions of higher k-mixture components  relatively have greater propensity to exhibit better fit than the lesser mixture component distributions (k < 3).

In the study, some bivariate distributions were developed from mixture model offspring, using the Independent (Product) distribution approach. These developments are categorized under the IID and IInD: where the Bivariate Exponential distribution, Bivariate Lindley distribution and Bivariate Juchez distribution are constructed as IIDs; and Bivariate Exponential-Lindley distribution, Bivariate Exponential-Juchez distribution and Bivariate Lindley-Juchez distribution as (IInDs). The properties of these distributions which involve: the shape of the bivariate PDFs, moments, moment generating function, mean, covariance and coefficient of correlation, maximum likelihood estimator, reliability analysis, renewal property and probability patterns; are studied across the distributions. Finally, under renewal properties, functions are derived which can model two-dimensional queuing and renewal processes, for events where the arrival and service times are dependent.

This paper examines the application of autoregressive integrated moving average (ARIMA) model and regression model with ARIMA errors for forecasting Nigeria’s GDP. The data used in this study are collected from the official website of World Bank for the period 1990-2019. A response variable (GDP) and four predictor variables are used for the study. The ARIMA model is fitted only to the response variable, while regression with ARIMA errors is fitted on the data as a whole. The Akaike Information Criterion Corrected (AICc) was used to select the best model among the selected ARIMA models, while the best model for forecasting GDP is selected using measures of forecast accuracy. The result showed that regression with ARIMA(2,0,1) errors is the best model for forecasting Nigeria’s GDP.