CLUSTER FAST DOUBLE BOOTSTRAP APPROACH WITH RANDOM EFFECT SPATIAL MODELING

Panel data is a combination of cross-sectional and time series data. Spatial panel analysis an to obtain information based on observations affected by the space or location effects. The effect effects spatial presented in form weighting. use regression provides number advantages, however, dependence test parameter estimators generated will be inaccurate when applied areas with small units. One method overcome problem unit size bootstrap method. This study used fast double (FDB) modeling poverty rate Flores islands. was sourced from BPS NTT Province website. results Hausman show that right model Random effect. concludes there islands tends SAR model. random R2 shows value 77.38 percent it does not meet assumption normality. Autoregressive Fast Double Bootstrap approach able explain diversity Island 99.83 fulfilling residual using FDB better than common panel.