Amemiya's Prediction Criterion penalizes R-squared more heavily than does
adjusted R-squared for each addition degree of freedom used on the
right-hand-side of the equation. The lower the better for this criterion.
$$((n + p) / (n - p))(1 - (R^2))$$
where n is the sample size, p is the number of predictors including the intercept and
R^2 is the coefficient of determination.
References
Amemiya, T. (1976). Selection of Regressors. Technical Report 225, Stanford University, Stanford, CA.
Judge, G. G., Griffiths, W. E., Hill, R. C., and Lee, T.-C. (1980). The Theory and Practice of Econometrics.
New York: John Wiley & Sons.
See Also
Other model selection criteria:
ols_aic(),
ols_fpe(),
ols_hsp(),
ols_mallows_cp(),
ols_msep(),
ols_sbc(),
ols_sbic()