A Probabilistic Model for Predicting Examination Performance: A Binary Time Series Regression Approach

I. U. MOFFAT

Abstract


In any examination result, performance is a function of several variables, which could be linear or nonlinear in dimension. Under approximately normal condition, every examination result is a Bernoulli trial with two unique and independent outcomes: a success and a failure. In this work, we examine the goodness - of - fit of the ordinary least squares regression with binary dependent variables (linear probability model) and the logistic regression in modeling and predicting examination performance. The degree examination results of 2012/2013 graduating class of the Department of Statistics were considered having reflected all the categories of performance in our examination grading system [viz; First class, Second class (Upper & Lower) divisions, Third class, and Pass]. The analysis revealed that the binary logistic regression is a better approach for modeling and predicting examination performance since most examination conditions are abnormal and nonlinear in dimension.


Keywords


Binary; binomial regression; logit; probability prediction; time series.

Full Text:

PDF

References


. Agresti, A. An Introduction to categorical data analysis, 2007, 423, Wiley Interscience.

. Long, J. S. Regression models for Categorical and Limited Dependent Variables. Advanced Quantitative Techniques in the Social Sciences, 1997, 7. Sage Publications: Thousand Oaks, CA.

. Powers, D., and Xie, Y. Statistical method for categorical data analysis Academic Press, 2000, San Diego, CA.

. Melton, J. Models for Binary dependent variables. IMT Institute of Advance studies Lecture Note, 2012.

. Han, M., and Michael, A. G. Multiple Regression Analysis and Mass Assessment: A Review of the Issues. The Appraisal Journal, 2001, 1, pp. 89 - 109.

. Hastie, T. and Tibshirani, R. Generalized additive models (2nd Edition), 1990, London: Chapman & Hall.

. Aldrich, J. H., and Nelson, F. D. Linear Probability, Logit, and Probit Models. John Herbert Aldrich Quantitative Applications in the Social Sciences, 1984, 45.

. Strano, M., and Colosimo, B. M. Logistic regression analysis for experimental determination of forming limit diagrams. International Journal of Machine Tools and Manufacture, 2006, pp. 46 49.


Refbacks

  • There are currently no refbacks.


 

 
  
 

 

  


About IJSBAR | Privacy PolicyTerms & Conditions | Contact Us | DisclaimerFAQs 

IJSBAR is published by (GSSRR).