Stepwise Multiple Regression Method to Forecast Fish Landing

There are six types of linear regression analyses that available in statistics which are simple linear regression, multiple linear regressions, logistic regression, ordinal regression, multinominal regression and descriminant analysis. Multiple linear regressions are the one of linear regression analyses that used to analyze the relationship between single response variable (dependent variable) with two or more controlled variables (independent variables). In this paper, stepwise multiple regression will use because this method is combination of forward selection and backward elimination method. The main objective in this paper is to select the suitable controlled variables in forecast fish landing. Data that has been used in this research were taken from Fisheries Annually Statistics of Department of Fisheries Malaysia. Then, the data will be analyzed by using Minitab 15 and SPSS 17.0.