第九讲 变量转换Transformation of Variables.pptVIP

SIB-BFSU ECONOMETRICS lECTURE 9 第九讲 变量转换Transformation of Variables LINEARITY AND NONLINEARITY LOGARITHMIC MODELSTHE DISTURBANCE TERM IN NONLINEAR MODELS Problem Set 6 1.The OECD data set on employment growth rates and GDP growth rates is listed in the table. Plot a scatter diagram and investigate whether a nonlinear specification might be superior to a linear one. Average rates of employment growth and GDP growth 1988-97 Employment GDP Australia 1.68 3.04 Austria 0.65 2.55 Belgium 0.34 2.16 Canada 1.17 2.03 Denmark 0.02 2.02 Finland -1.06 1.78 France 0.28 2.08 Germany 0.08 2.71 Greece 0.87 2.08 Iceland -0.13 1.54 Italy -0.30 1.68 Japan 1.06 2.81 Employment GDP Korea 2.57 7.73 Luxembourg 3.02 5.64 Netherlands 1.88 2.86 New Zealand 0.91 2.01 Norway 0.36 2.98 Portugal 0.33 2.79 Spain 0.89 2.60 Sweden -0.94 1.17 Switzerland 0.79 1.15 Turkey 2.02 4.18 U.K 0.66 1.97 U.S.A 1.53 2.46 Problem Set 6 2. Using your data set, regress the logarithm of earnings on S and ASVABC. Interpret the regression results and perform appropriate tests. THE DISTURBANCE TERM IN NONLINEAR MODELS v f(v) Besides satisfying the Gauss-Markov conditions, we need u to be normally distributed if we are to perform t tests and F tests. THE DISTURBANCE TERM IN NONLINEAR MODELS v f(v) This will be the case if v has a lognormal distribution