ANALYSIS OF THE MOST IMPORTANT CAUSES OF LIVER CANCER WITH A PRACTICAL APPLICATION USING THE LOGISTIC REGRESSION METHOD

In order to determine the most important causes of liver cancer, the researcher had to conduct a statistical analysis using a statistical model suitable for this type of data, which is medical data.The logistic regression model was used to determine the parameters, and with the help of the ready-made statistical application (SPSS), the data analysis process was completed and the research sample was It consists of (120 observations) patients lying in the Medical City Hospital in Bab al-Muadham / Baghdad who are suspected of being infected with the disease.It turned out that of these patients, (93 patients) were infected with the disease and ( 27patients) were not infected with the disease.This sample was divided into two parts, women with a number of (64 observations) and men (56 observations).Fifteen independent variables (x) were used, which are the variables studied in the research and have a direct impact on the response variable (the dependent variable (y)).When conducting the statistical analysis, we obtained the most important results, which is that the calculated value of the Wald test (WALD = 32.006),which is greater than its tabulated value, which is equal to (WALD = 30.09).Therefore, the null hypothesis is rejected and we accept the alternative hypothesis.This means that the relationship is significant between the independent variables studied (Xi) and the dependent variable (y), meaning that there is a direct effect of these studied variables on the response variable (incidence of liver cancer).The results also showed that the value of the Cox test (cox = 0.648), which is a test corresponding to the chi-square test, and its interpretation means that its value (0.648) was achieved.It is interpreted from the logistic regression model, which is a very good percentage and is calculated based on the strength of the research and the strength of the results of the variables studied, and the rest is placed in the error term, meaning that there are other independent variables that have an impact on the response variable (y).(It was not studied in this research, and the results also showed that the independent variables that have the most influence on the response variable are primarily the variable (X14), which represents the incidence of another cancer such as lung