2022, Vol. 3, Issue 2, Part B
A comparison of ordinary least squares and weighted regression for road accident causality models
Author(s): Ajibode IA and Agbolade Olumuyiwa O
Abstract: The examination of data on road accident fatalities was used in this study to determine which of the two estimators, the Weighted Least Square Estimator (WLS) and the Ordinary Least Square Estimator (OLS), is most effective. The Federal Road Safety Corps' official website provided statistics for this study's principal data source for a 50-year span, from 1972 to 2021, making it secondary in nature. In order to estimate the relevant parameters, the Ordinary Least Squares (OLS) and Weighted Least Squares (WLS) estimators were utilised. Numerous assessment measures were used to compare these estimators, including R-squared, Adjusted R-squared, Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Mean Squared Error (MSE). The evidence indicates that, when these particular qualities are taken into account, WLS performs better than OLS. The conclusion that the weighted least squares (WLS) estimator is better than the ordinary least squares (OLS) estimate in the setting of traffic data is therefore logical. So, before putting the best model into practise, it is crucial to identify it.
Pages: 134-137 | Views: 426 | Downloads: 154
Download Full Article: Click Here
How to cite this article:
Ajibode IA and Agbolade Olumuyiwa O. A comparison of ordinary least squares and weighted regression for road accident causality models. Journal of Mathematical Problems, Equations and Statistics. 2022; 3(2): 134-137.
