Figure 3 hausman FEM REM Coefficients (b) (B) FEM REM (b-B) Difference sqrt (diag (V_b-V_3)) S.E. stokpr end .075275 .0064233 .0692905 .0013278 .0059845 .0050955 .0032798 .0016491 b = consistent under Ho and Ha; obtained from streg B - inconsistent under Ha, efficient under Ho; obtained from streg Test: HO: difference in coefficients not systematic chi2 (2) - (b-B)'[(V_b-V_B)^(-1)] (b-B) 9.82 Prob>chi2 = 0.0074 Consider the Hausman test results in Figure 3. Which conclusion can you draw? At the 1% significance level REM is preferred to FEM None of the above The LSDV model explains 0.0074% of the variation in the dependent variable There is no collinearity bias present in the regression equation at the 5% significance level Heteroskedasticity is present at the 1% significance level