Many research agencies predict the result of election using various methods before election. Basic purpose of survey is to forecast the election result. One of the major problems to forecast election, especially based on survey, is non-response. We ma...
Many research agencies predict the result of election using various methods before election. Basic purpose of survey is to forecast the election result. One of the major problems to forecast election, especially based on survey, is non-response. We may have different forecasting results depend on method of non-response imputation. In this research, we consider a model based method of non-response imputation. An assumption of the non-response mechanism is very important precondition to forecast the accurate results. In this research, we compared the accuracy of prediction and assumption of non-response data by result of presidential election exit poll. We consider maximum likelihood estimation based on EM algorithm to handle assumption of the model of non-response data. We also consider modified within precinct error which Bautista(2007) proposed to compare the predict result.