Dear Statalisters,
I wanted to validate the predictive accuracy of my estimates but could not
find any relevant test in Stata. I was particularly looking for tests like
Theil stats to validate
the predicted values vs actual but couldnt find it. I had the following
results (sse below) after predict and I just want to know how good the
predictions are. Is there any way in Stata that I can do some validation
how good the regression (-xtpcse-) estimates could predict the dependent
variable, other than the graphs?
Thanks.
Cristina
predict yhat
sort time label
table time label, c(mean yield mean yhat)
------------------------------
| label
time | 0 1
----------+-------------------
1 | 311.2743 306.1385
| 305.026 311.0978
|
2 | 315.9229 299.9346
| 311.8012 319.4947
|
3 | 325.5143 328.4192
| 318.5678 325.8331
|
4 | 302.1857 343.7077
| 324.1809 334.4826
|
5 | 336.5 336.9846
| 330.8871 340.565
|
6 |
| 337.6405 347.3184
|
7 |
| 344.394 354.0718
|
8 |
| 351.1474 360.8253
|
9 |
| 357.9008 367.5787
|
10 |
| 364.6542 374.3321
|
11 |
| 371.4077 381.0856
|
12 |
| 378.1611 387.839
|
13 |
| 384.9145 394.5924
|
14 |
| 391.668 401.3459
|
15 |
| 398.4214 408.0993
|
16 |
| 405.1748 414.8527
|
17 |
| 411.9283 421.6062
|
18 |
| 418.6817 428.3596
|
19 |
| 425.4351 435.113
|
20 |
| 432.1886 441.8665
------------------------------
.
.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
