Statistical Analysis: Predictions
Retained estimation results (all accessible by name in subsequent operations)
- Predictions produced by all models
- Retain as a variable in the data set
- List with residuals and confidence intervals
- Extrapolate to out of sample observations
Example
Results shown will differ by model. The following shows the analysis of fit and listing of predictions for a binomial logit model.
+----------------------------------------+
| Fit Measures for Binomial Choice Model |
| Logit model for variable MODE |
+----------------------------------------+
| Proportions P0= .750000 P1= .250000 |
| N = 100 N0= 75 N1= 25 |
| LogL = -46.20450 LogL0 = -56.2335 |
| Estrella = 1-(L/L0)^(-2L0/n) = .19822 |
+----------------------------------------+
| Efron | McFadden | Ben./Lerman |
| .21782 | .17835 | .70349 |
| Cramer | Veall/Zim. | Rsqrd_ML |
| .20930 | .31562 | .18174 |
+----------------------------------------+
| Information Akaike I.C. Schwarz I.C. |
| Criteria .98409 106.22452 |
+----------------------------------------+
Frequencies of actual & predicted outcomes
Predicted outcome has maximum probability.
Threshold value for predicting Y=1 = .5000
Predicted
------ ---------- + -----
Actual 0 1 | Total
------ ---------- + -----
0 63 12 | 75
1 12 13 | 25
------ ---------- + -----
Total 75 25 | 100
Predicted Values
Obs. Observed Y Predicted Y Residual x(i)b Pr[Y=1]
1 .00000 .00000 .0000 -2.9586 .0493
2 .00000 .00000 .0000 -1.3397 .2076
3 .00000 .00000 .0000 -1.3859 .2001
4 1.0000 1.0000 .0000 .2351 .5585
5 .00000 .00000 .0000 -2.7272 .0614
6 .00000 .00000 .0000 -1.8030 .1415
7 .00000 .00000 .0000 -2.2193 .0980
8 1.0000 1.0000 .0000 .2340 .5582
9 .00000 .00000 .0000 -2.9617 .0492
10 .00000 .00000 .0000 -1.3463 .2065