use "E:\stata\sample\chap16\案例16.1.dta", clear

. cd E:\STATA12.0\ado\xtptm
E:\STATA12.0\ado\xtptm

. encode diqu,gen(region)

. xtset region year
panel variable:region (strongly balanced)
time variable:year, 2008 to 2012
delta:1 unit

. xtptm profit,rx(sale) thrvar(cost) iters(2000) grid(200)

================ Fundamental information: ===================

Number of Regime independent variables: 0
Number of Regime dependent variables: 1
Number of individuals in panel: 20
Number of periods in panel: 5

================ Ordinary Fixed effect regression: =========

Sum of Squared Residuls: 1.1015
Stardard error of regression: 0.1181
Regression Result(ordinary standard error):
----------------------------------------------------
| Coef Std t Prob
----+-----------------------------------------------
1 | 0.0006 0.0005 1.3678 0.1752 （这里为什么只有一个系数？）
----------------------------------------------------
Regression Result(Robust standard error): （为什么又多了个稳健标准误？有什么作用？）
-----------------------------------------------------
| Coef Std_Robust t Prob
----+------------------------------------------------
1 | 0.0006 0.0004 1.6849 0.0960
-----------------------------------------------------

================ Single threshold regession: =================== 单门槛回归吗？单是只一个门槛变量还是指一个门槛值？

Minimized Sum of Squared Residuals: 0.8822
Standard error of residuals: 0.1064
Threshold estimator: 11.3611
95% conf. intv. of threshold: 11.3274 11.3836
LR Critical value to test gamma=gamma0: 7.3523 （请问这行结果怎么解读？）
Threshold regression(Ordinary Std. Error):
----------------------------------------------------
| Coef Std t prob
----+-----------------------------------------------
1 | -0.0001 0.0005 -0.2480 0.8048
2 | 0.0006 0.0004 1.3794 0.1717（系数分别对应到什么变量阿，有常变量吗？）
----------------------------------------------------
Threshold regression(Robust Std. Error):
-----------------------------------------------------
| Coef Std_Robust t prob
----+------------------------------------------------
1 | -0.0001 0.0004 -0.3052 0.7611
2 | 0.0006 0.0003 1.6497 0.1030
-----------------------------------------------------
Note: Critcal (Inverse CDF of LR stat): -2*ln(1-sqrt(1-alpha))
Ho: No threshold; Ha: Single threshold
Number of bootstrap:2000
F-stat & Prob: 19.3819 0.0000
F-critical value of 90% 95% 99%:
1
+---------------+
1 |2.859457574|
2 |4.282186037|
3 |7.315390308|
+---------------+
Thresholds in single model:
11.36112623

============== Descrpitive statistic in each regime: =========== （大神帮忙解释下以下结果）

Descriptive statistics of y-X-THR at regime :1
----------------------------------------------------------------
| Mean Std Min Max Count
----+-----------------------------------------------------------
1 | 10.0823 0.3656 9.2360 10.5545 52
2 | 217.1043 18.9613 138.0000 230.2526 52
3 | 11.0819 0.1916 10.7579 11.3516 52
----------------------------------------------------------------
Descriptive statistics of y-X-THR at regime :2
----------------------------------------------------------------
| Mean Std Min Max Count
----+-----------------------------------------------------------
1 | 10.6126 0.9030 9.6125 12.4765 48
2 | 233.6324 41.5348 89.0000 346.0000 48
3 | 11.9187 0.6118 11.3667 13.2804 48
----------------------------------------------------------------
LR and Thresholds series are stored in Stata matrix: LR#
You can observe the scatter plot by: _matplot LR#, columns(1 2)

还有 最后的门槛值是确定了只有一个吗？还是要继续做regime(2),regime(3)阿？