Using a Method(BYC) to Estimate the Parameters of a Circular Logistic Regression Model with a Practical Application"

Abstract

The research included a comparison between the circular maximum likelihood method and the robust circular (BYC) method to estimate the parameters of the circular logistic regression model represented by the Maximum Likelihood Method. In the case of the presence of outliers in the data, robust estimation methods were relied upon to estimate the parameters of the circular logistic regression model, namely the (BY) method, using the mean square error (MSE) criterion, conducting a Monte-Carlo simulation for small, medium and large regression samples and with different pollution ratios to study the behavior of the estimation methods under study. It was concluded that the (BYC) method is better than the rest of the estimation methods, as it achieved the lowest comparison criteria, but this method was close to the regular maximum likelihood method in the case that the larger the sample size, the greater the preference of the robust circular estimation methods. As for the applied aspect of this study, the (BYC) method was used to estimate the parameters of the circular logistic regression model. The model and its application to real data representing a simple random sample of (120) circular variables, namely the effect of wind direction on rainfall.