1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
|
/*
Replicate three of the models estimated by Baltagi
(Econometric Analysis of Panel Data, 3e, chapter 7),
based on North Carolina crime panel data from Cornwell
and Trumbull
*/
set messages off
include ivpanel.gfn
open nc_crime.gdt --quiet
# primary regressors
list X = lprbarr lprbconv lprbpris lavgsen lpolpc ldensity \
lwcon lwtuc lwtrd lwfir lwser lwmfg lwfed lwsta lwloc lpctymle
# time-invariant regressors
list invariant = lpctmin west central urban
# time dummies
list TD = d82 d83 d84 d85 d86 d87
# endogenous regressors
list Endo = lpolpc lprbarr
# instruments
list Inst = ltaxpc lmix
# fixed effects: can use time dummies
list FX = X TD
list FZ = FX - Endo Inst
# Fixed-effects TSLS
ivpanel(lcrmrte, FX, FZ, 1)
# between: exclude time dummies but add time-invariant series
list BX = X invariant
list BZ = BX - Endo Inst
# Between TSLS
ivpanel(lcrmrte, BX, BZ, 2)
# random effects (G2SLS): use all regressors
list RX = X invariant TD
list RZ = RX - Endo Inst
ivpanel(lcrmrte, RX, RZ, 3)
|
