Tag Archives: Logistic regression

Predicting Student University Admission Using Logistic Regression (Published)

The primary purpose is to discuss the prediction of student admission to university based on numerous factors and using logistic regression. Many prospective students apply for Master’s programs. The admission decision depends on criteria within the particular college or degree program. The independent variables in this study will be measured statistically to predict graduate school admission. Exploration and data analysis, if successful, would allow predictive models to allow better prioritization of the applicants screening process to Master’s degree programme which in turn provides the admission to the right candidates.

Keywords: Logistic regression, college admission, data analytics, predictive analysis

Analysis of the Viability of Poultry Enterprises in Imo State, Nigeria (Published)

Against the background of the ban on the importation of poultry products, and growing interest in the local Poultry production, this study looked at the viability of poultry enterprises in Imo State, Nigeria. It specifically analyzed the profit level of these enterprises and estimated their viability using the Benefit-Cost Ratio. Logistic regression technique was used to estimate the determinants of viability of the poultry enterprises in the area.  Sixty Poultry enterprises were randomly selected from Owerri Agricultural Zone of the State and their entrepreneurs interviewed. Data were collected by means of structured questionnaire.  Descriptive statistics, net returns model and Benefit-Cost Ratio were analytical tools used to achieve the objectives of the study. The results showed that males dominated the poultry business in the study area and the average age of the entrepreneurs was 44 years. Majority of the entrepreneurs attained some level of formal education and had average number of employees of about 6 persons. The poultry enterprises in the area were found to be profitable and viable with profit level of N188163.86 and BCR of 1.68.   Sex, marital status, age, farming experience, facility size, returns and access to veterinary services were the determinants of viability of the poultry enterprises in the area. Age of entrepreneur, marital status, facility size, access to veterinary services and returns positively influenced the viability of poultry enterprises in the area. Years of experience of the entrepreneur negatively influenced viability of the poultry enterprises in the study area. The study recommended the encouragement of more females to venture into poultry production. It also advocated the provision of efficiency and easily accessible veterinary services for the poultry farmers in the area. 

Keywords: Imo State, Logistic regression, Poultry Enterprises, Viability, profit

Analysing Dependent Variables with Multiple Surrogates in Financial Performance Research (Published)

Accounting and finance-based researchers often use multiple surrogates to capture the properties of a dependent variable (DV) when studying its predictive relationship with predictors. This often fail to directly connect the study results with the major objective of the research. This paper compares the existing practice with a plausible and less complicated alternative. Using logistic regression, the study converted the a priori expectations of 30 Ph.D research theses in finance and accounting with four dependent surrogates into a probabilistic log values and compared them with the individual surrogate performance on the one hand and the surrogates geometric mean on the other hand. While the geometric mean revealed close connection with the theses’ probabilistic expectations (β = .278, t(30) = .695, R2 = .077, p > .10), the individual surrogates results showed singular and combined significant differences with the theses’ a priori expectations (Adj. R2 = .0291, F(4, 25) = 22.598, p < .05). The paper recommends unifying multivariate DVs with geometric means for better conclusion in financial performance relational studies.     

Keywords: Logistic regression, Performance, dependent variable, proxy, surrogate

Evaluation of Factors Influencing Switching Behaviour by Ghana Commercial Bank Customers (Published)

This study seeks to determine the factors which influence the switching behaviour of customers of Ghana Commercial Bank, Limited in Ho. A descriptive, cross-sectional survey was conducted among 350 purposively selected individual customers. Logistic regression analysis was used to identify the predictors of switching intentions among customers. Results show that four factors; X1 (High transaction fee), X4 (Attractiveness of alternatives), X7 (Inconvenience of bank location) and X9 (Inability to respond to system failure quickly) were statistically significant in the prediction of customer switching with a predicted switching rate of 82.29%. It is suggested there is a need for banks to review their bank charges or transaction fees in the banking sector since high transaction fees have an impact on customer switching behaviour. Also, management should establish more branches in the same township since customers switch in the inception of convenience in the services and location of the banks. Finally, banks should regularly update their system and also employed welled trained staff who will respond to system failure quickly. In addition, they should strive to provide the greatest possible customer satisfaction and convenience them that they have greater customer satisfaction than competitive banks.

Keywords: Customers, Logistic regression, Switching Behaviour


The attack on a computer system with the intention of finding security weaknesses are becoming increasingly frequent and evermore sophisticated, potentially gaining access to it, its functionality and data. Organizations wishing to ensure security of their systems may look towards adopting appropriate tests to protect themselves against potential security breaches. One such test is to hire the services of penetration testers (or “pen-tester”) to find vulnerabilities present in the case study for “Cairo Cleaning and Beautification Agency”, and provide recommendations as to how best to mitigate such risks. By using series of the standards built on the application of data mining methods specifically decision tress model, Logistic regression, association rules model, Bayesian network for making reference penetration testers. This paper discusses the definition and role of the modern pen-tester and summaries current standards and professional qualifications. The paper further identifies issues arising from pen-testers; their motivation is to improve security.

Keywords: Bayesian network, Logistic regression, Penetration testing, association rules model, cyber security, vulnerability assessments decision tress model