Solving the probability regression models by using r programming environment : case study for patients epilepsy in Basra

Abstract EN

The technology for dealing with accurate calculations progresses very quickly .

and be a major impetus for scientific progress.

and the perfect mastery of R in higher levels is necessary for many applications .

The R statistical programming environment provides an ideal stage to conduct the study because of its powerful programming capability , graphics and a comprehensive set of statistical functions , it contains more than 11164 packages .

The aim is to study the phenomenon of epilepsy and to identify the most important factors affecting patients with epilepsy in the province of Basra.

And the cause of this disease of social and economic effects on the patient and his family.

and also this study is to solve the probability regression models represented by probit, tobit and logit ,as well as identifying the variables of the study (sex and age) that affect or increase the incidence epilepsy for the data of 2296 patients in Basra.

The study design , is the represent by estimating the three probable regression models by probit, tobit and logit for the effect of sex and age factors on the risk of developing epilepsy ,and the tools used is to make statistic program ,then we find the R programming language was give the good and correct answer, Then the statistical analysis using R language in which the written orders are implemented directly without the need to build a complete program to implement the programming statements using the regression function for each probability regression models (Probit, Tobit and Logit) and the results showed that the regression of the probabilistic functions is that both the age factor and the sex factor have a significant effect in the case of the disease in terms of being inpatient or outpatient.

This is confirmed by the Wold test.

and Also the conclusion shows that functions represented the best representation, which was confirmed by the coefficient of selection as it reached 0.96.