
Re: Correct way to normalize an rmsdbased distance metric used in repeated trials of pairs
Posted:
Apr 20, 2012 3:43 PM


I don't understand the results for fold a1, length 8392. How can the coefficient for predictor 1 be > 1 ?
0,0,27,24 0,1,46,152 1,0,30,304 1,1,26,137
Descriptives...
129 cases have Y=0; 617 cases have Y=1.
Variable Avg SD 1 0.6662 0.4716 2 0.4839 0.4997
Iteration History... 2 Log Likelihood = 687.0506 (Null Model) 2 Log Likelihood = 651.8668 2 Log Likelihood = 650.2258 2 Log Likelihood = 650.2243 2 Log Likelihood = 650.2243 (Converged)
Overall Model Fit... Chi Square= 36.8263; df=2; p= 0.0000
Coefficients and Standard Errors... Variable Coeff. StdErr p 1 1.2937 0.2285 0.0000 2 0.2297 0.2292 0.3162 Intercept 0.6991
Odds Ratios and 95% Confidence Intervals... Variable O.R. Low  High 1 3.6463 2.3299 5.7063
2 1.2582 0.8029 1.9717
X1 X2 n0 n1 Calc Prob 0.0000 0.0000 27 24 0.6680 0.0000 1.0000 46 152 0.7168 1.0000 0.0000 30 304 0.8800 1.0000 1.0000 26 137 0.9023

