PEMODELAN TINGKAT KEMISKINAN DI LIMA PROVINSI INDONESIA DENGAN REGRESI DATA PANEL

In conducting a study on poverty levels in five provinces in Indonesia, is not enough to make the observation of the units at the same time, but also observe the behaviorof the unit at different time periods during a period of six years, which is 2008 to 2013. It required data that is a combination of data cross section and time series, also called panel data. Panel data regression is used to model the effect of economic growth, minimum wage and unemployment against Poverty Level. Estimation of panel data regression model will be performed by three approaches, namely common effect, fixed effect and random effect approach. The results of the test can be concluded that the more precise the data using fixed effects model (fixed effect) where the Economic Growth and Wages miminum negatively impact the levelof poverty, as well as the positive impact on the unemployment rate Poverty Level Keywords: Panel Data Regression,Common Effect, Fixed Effect, Random Effect