例如，在plm包，pgmm程序里，如想度量出时间虚拟变量的效应，该如何实现？

谢谢，需要重新编写还是套用程序？

z1<-pgmm(dynformula(log(emp)~log(wage)+log(capital)+log(output),list(2,1,0,1)),data=EmplUK,effect="twoways",model="twosteps", gmm.inst=~log(emp),lag.gmm=list(c(2,99)))

summary(z1,robust=TRUE)

 

Twoways effects Two steps model

 

Call:

pgmm(formula = log(emp) ~ lag(log(emp), 1) + lag(log(emp), 2) +

    log(wage) + lag(log(wage), 1) + log(capital) + log(output) +

    lag(log(output), 1), data = EmplUK, effect = "twoways", model = "twosteps",

    gmm.inst = ~log(emp), lag.gmm = list(c(2, 99)))

 

Unbalanced Panel: n=140, T=7-9, N=1031

 

Number of Observations Used:  611

 

Residuals

      Min.    1st Qu.     Median       Mean    3rd Qu.       Max.

-0.6191000 -0.0494800 -0.0004565 -0.0001841  0.0533500  0.6410000

 

Coefficients

                     Estimate Std. Error z-value  Pr(>|z|)   

lag(log(emp), 1)     0.474151   0.185398  2.5575 0.0105437 * 

lag(log(emp), 2)    -0.052967   0.051749 -1.0235 0.3060506    

log(wage)           -0.513205   0.145565 -3.5256 0.0004225 ***

lag(log(wage), 1)    0.224640   0.141950  1.5825 0.1135279   

log(capital)         0.292723   0.062627  4.6741 2.953e-06 ***

log(output)          0.609775   0.156263  3.9022 9.530e-05 ***

lag(log(output), 1) -0.446373   0.217302 -2.0542 0.0399605 * 

---

Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

 

Sargan Test: chisq(25) = 30.11247 (p.value=0.22011)

Autocorrelation test (1): normal = -1.538450 (p.value=0.061969)

Autocorrelation test (2): normal = -0.2796829 (p.value=0.38986)

Wald test for coefficients: chisq(7) = 142.0353 (p.value=< 2.22e-16)

Wald test for time dummies: chisq(6) = 16.97046 (p.value=0.0093924)