Logistic Regression without Intercept

Guoping Zeng

Asian Journal of Probability and Statistics, Page 30-42
DOI: 10.9734/ajpas/2022/v17i130414

Logistic regression is a popular statistic modelling algorithm in predicting a binary outcome. Although logistic regression almost always has an intercept, logistic regression without intercept is sometimes appropriate or even necessary. However, logistic regression without intercept has rarely been discussed other than being used explicitly or implicitly. In this paper, we aim at filling this gap by systematically studying logistic regression without intercept. Specifically, we study the 4 most important aspects of logistic regression: (1) Maximum Likelihood Estimate, (2) data configuration (complete separation, quasi-complete separation and overlap) to categorize the existence and uniqueness of maximum likelihood estimate, (3) multicollinearity, and (4) monotonic transformations of independent variables. We adopt an extensional method in that we first present results for logistic regression with intercept and then extend the results to the case of without intercept. Our numerical examples further compare logistic regression with intercept and without intercept.

Spatial and Temporal Simulation of Typhoid Fever Transmission in Yobe State

Babagana Modu, Sheriff Wakil, Umar Y. Madaki, Audu M. Mabu

Asian Journal of Probability and Statistics, Page 1-11
DOI: 10.9734/ajpas/2022/v16i530411

Aims / Objectives: Typhoid fever is a threat to human race and perhaps not much research is conducted towards mitigating it menace in Yobe State. A classical epidemic model SIR is deployed into GLEaMviz software to simulate typhoid spread and spatially analysed the trend.

Study Design: Computational modeling and simulation.

Place and Duration of Study: Computational Laboratory, Department of Mathematics and Statistics Yobe State University, Damaturu, Nigeria. The duration of the study is between May 2021 and December 2021.

Methodology: SIR epidemic model was used to simulate typhoid spread and time series model was explored to investigate the disease trend.

Results: The model predicts mild seasonal fluctuations in the trend which coincides with rainy season. The agents causing the disease transmission is possibly being transported through flowing water.

Conclusion: A mild seasonality is present in the fluctuations of the trend of typhoid, hence the pattern shows strong evidence of perennial tendency with likelihood of high cases during rainy season. Further work is needed to validate this findings by using real data.

Arima Model for Forecasting of Monthly Rainfall and Temperature in the Lake Victoria Basin

Oryiema Robert

Asian Journal of Probability and Statistics, Page 12-22
DOI: 10.9734/ajpas/2022/v17i130412

The rise in global temperature which is global warming has lead to erratic and disruptive weather pattern in several regions of the world including the area surrounding the Lake Victoria basin. Likewise, economic activities associated with Lake Victoria and its Basin such as agriculture, fishing, mining and transportation are significantly affected by this climatic changes. The primary cause of negative impact that stems from this changes is lack of reliable information that can be used to predict and address the climatic variations within the basin. The objective of this research is to identify a suitable time series model that can be used to analyse and predict this weather variations and pattern around the Lake Victoria basin. This research uses Box jenskin methodology to build ARIMA(2,0,1) model for rainfall pattern around Lake Victoria basin. The data is obtained from three Kenya Meteorological Department weather stations as secondary data from the years 2008 to 2014. In this research, data from the years 2008 to 2010 was used to estimate the values for the years 2011 to 2013. The relationship from the research showed a strong positive relationship which indicates high level of accuracy on predictability by the model.

A Statistical Analysis on the Effect of Bad Health Habits in two Continents (Africa and Europe)

Ogonna A. Ndubuisi, Uchenna P. Ogoke

Asian Journal of Probability and Statistics, Page 23-29
DOI: 10.9734/ajpas/2022/v17i130413

This research work focuses on determining the difference between the health habits of countries in Africa and Europe, especially in females. It is crucial because it could help enlighten women on the dangers of bad health habits. Multivariate Hotelling T- square test is adopted to determine the significant difference between the two continents, Africa and Europe, having Cancer deaths caused by alcohol consumption, smoking prevalence, and Obesity prevalence as the variables and the correlation between the variables. The result showed that there is indeed a significant difference between the bad health habits in the two continents in the females. Correlation analysis was also carried out to determine the relationship between the variables, and the results showed that the relationship between the variables was little. More methods were adopted using comparison of Rate between the means, and the sum of the same variables, to determine which continent was more affected by the bad health habits, and it was figured out to be Europe.

On the Comparison of Ordinary Least Squares and Quantile Regression with Nigerian Financial Data on Trade Balance, Foreign Inflow and Imports

Ogoke Uchenna Petronilla

Asian Journal of Probability and Statistics, Page 43-52
DOI: 10.9734/ajpas/2022/v17i130415

This research work compared the Ordinary Least Squares Regression and Quantile Regression models, as well as the differences between them thereby examining the compared models in terms of goodness of fit statistic and also recommend a suitable fit model for the data collected. These methods were applied to the Nigerian Financial Data on Trade Balance, Foreign Inflow and Imports on the Nigerian Gross Domestic Product. The results via this study, shows that the influence of Trade Balance, Foreign Inflow and Imports vary on the Nigerian Gross Domestic Product depending on the quantile one is looking at. This study recommends the robustness and stability of the Quantile regression model considered as an alternative to the Ordinary Least Square Model.