C. VAN DE VOORDE ET AL 寸∞9-n∞∽ ccs en 3E g 百 五8 E ≥。多 目目 9 百直 Copyright a 2001 John Wiley Sons, Ltd Health Econ.10:457-471(2001)468 C. VAN DE VOORDE ET AL. Copyright © 2001 John Wiley & Sons, Ltd. Health Econ. 10: 457–471 (2001) Table 2. Fixed effects (WLSDV) estimation results for the differences model (2) t-statistic Difference Robust t-statistic Adjusted White Adjusted Difference in Robust S.E.a Durbin – (price) ln in trends R2 S.E.a testb Watsonc GP home visits per person −0.012 0.002 −6.0 0.9997 −0.281 1.881 1.665 ln(rate of gen pop) −ln(rate of 0.034 −8.3 low-income WOPI) ln(rate of WOPI) −ln(rate of −0.6 −0.000 0.001 −0.1 0.9968 0.506 1.364 −0.018 0.030 low-income WOPI) GP office visits per person −0.001 0.001 −0.8 0.9579 1.094 −13.5 1.217 0.009 −0.118 ln(rate of gen pop) −ln(rate of low-income WOPI) 0.015 0.001 ln(rate of WOPI) −ln(rate of 13.5 0.9515 0.857 1.684 −0.031 0.023 −1.3 low-income WOPI) Specialist visits per person −7.0 ln(rate of gen pop) −ln(rate of −0.007 0.001 −8.9 0.9576 0.792 0.997 −0.103 0.015 low-income WOPI) 0.003 0.001 2.2 0.9306 0.942 0.027 2.012 −0.042 −1.5 ln(rate of WOPI) −ln(rate of low-income WOPI) a Robust standard error estimates using Arellano’s method [17]. b White test statistic 1.96 indicates stat significant heteroscedasticity at 95% level. c All adjusted Durbin–Watson statistics indicate positive autocorrelation, see the table in [16].