Bias Not Linear Prediction Method The Best for Small Area on Empirical Suspect Spending Per Capita (Case: South Sulawesi Province District Gowa)

Indirect estimation is the estimation of parameters that use additional information about the same parameters in a small area. The combination of the basic assumption of random effects and fixed effects forms a mixed influence model. In this study, the data used are secondary data from the SUSENAS and Regression methods. The value of Average Root Mean Square Error (ARSME) estimator in the EBLUP estimator is smaller when compared to the direct estimator and simple linear regression. So it can be said that the EBLUP estimator is better than the direct estimator of per household at the District level in Gowa Regency based on SUSENAS data for 2017.