初学者,最近在做模糊断点分析,研究退休对各种营养元素的影响,运用模糊断点分析下来,比如退休对碳水化合物无显著断点影响,lwald值p值为0.636,不显著。但是如果直接回归退休虚拟变量对碳水化合物的影响,表现出显著相关关系。断点回归和面板回归做出来的结果就明显不一样。下面是RD回归和IV面板的结果,请大神帮帮忙
rd d3carbo retirement1 age11,mbw(100)
Three variables specified; jump in treatment
at Z=0 will be estimated. Local Wald Estimate
is the ratio of jump in outcome to jump in treatment.
Assignment variable Z is age11
Treatment variable X_T is retirement1
Outcome variable y is d3carbo
Estimating for bandwidth 8.329601996943215
------------------------------------------------------------------------------
d3carbo | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
numer | 4.341706 9.098546 0.48 0.633 -13.49112 22.17453
denom | .4008749 .0350789 11.43 0.000 .3321215 .4696283
lwald | 10.83058 22.86314 0.47 0.636 -33.98036 55.64151
------------------------------------------------------------------------------
. xtivreg d3kcal (retirement1=age1) education marriage nutritionknowledge
G2SLS random-effects IV regression Number of obs = 2935
Group variable: idind Number of groups = 1125
R-sq: within = 0.0313 Obs per group: min = 1
between = 0.0059 avg = 2.6
overall = 0.0129 max = 7
Wald chi2(4) = 52.05
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
------------------------------------------------------------------------------
d3kcal | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
retirement1 | -208.3232 32.98659 -6.32 0.000 -272.9757 -143.6706
education | -12.92452 8.670695 -1.49 0.136 -29.91877 4.06973
marriage | -6.426839 31.11156 -0.21 0.836 -67.40437 54.5507
nutritionk~e | -81.3054 31.47609 -2.58 0.010 -142.9974 -19.6134
_cons | 2474.104 70.12467 35.28 0.000 2336.662 2611.546