大家好,我想请教一下关于截面数据的门限效应回归结果的一个问题:
我根据Hansen(2000)文献提供的数据和代码,用的thresholdtest 和thresholdreg命令,跑出来的结果是下面的样子:
命令:thresholdreg GDP_Growth log_GDP InvGDP Pop_Growth school, q(literacy) h(1)
结果:
Threshold Estimation
______________________________________________________________________
Threshold Estimate: 45
.95 Confidence Iterval: [19,57]
Sum of Squared Errors 6.19824886
Residual Variance: .091150719
Joint R-Squared: .619872813
Heteroskedasticity Test (p-value):.551557
______________________________________________________________________
Regime1 q<=45
Parameter Estimates
______________________________________________________________________
Independent Variables Estimate St Error
______________________________________________________________________
Intercept 2.09228142 1.87019128
log_GDP -.116316839 .164871236
InvGDP .172642306 .21614631
Pop_Growth -.390217638 .516128646
school .45253325 .116828949
.95 Confidence Regions for Parameters.
Independent Variables Low High
______________________________________________________________________
Intercept -1.85384855 6.16104093
log_GDP -.470873901 .344718316
InvGDP -.318180882 .655483468
Pop_Growth -1.59323556 .868347526
school .195380576 .74597169
Observations: 30
Degrees of Freedom: 25
Sum of Squared Errors: 2.60999433
Residual Variance: .104399773
R-squared: .578036575
Regime2 q>45
Parameter Estimates
______________________________________________________________________
Independent Variables Estimate St Error
______________________________________________________________________
Intercept 4.31048461 .965159759
log_GDP -.395026857 .061027541
InvGDP .833639859 .139369899
Pop_Growth -.418009389 .269566721
school .094577957 .13489238
.95 Confidence Regions for Parameters.
Independent Variables Low High
______________________________________________________________________
Intercept 1.58189222 6.22915676
log_GDP -.535433992 -.249519463
InvGDP .427114365 1.13187913
Pop_Growth -1.11334431 .11458144
school -.344314463 .402081447
Observations: 48
Degrees of Freedom: 43
Sum of Squared Errors: 3.58825453
Residual Variance: .08344778
R-squared: .582117454
我想请问一下,为什么作为门槛的变量没有参数估计值?只要其他控制变量才有参数估计呢?


雷达卡







京公网安备 11010802022788号







