论文信息 - On use of predictive probabilistic estimates for selecting best decision rules in the course of a search

On use of predictive probabilistic estimates for selecting best decision rules in the course of a search

The problem of how to find the 'best' decision rule in the course of a search with the help of analysis of sample set is considered. Specifically the problem of selecting of best subset of regressors is highlighted. The concepts of predictive probabilistic estimate (PPE), decomposition of a search process on stages, ensemble of noise functions and reference probability distribution on it are introduced and discussed. A Monte Carlo procedure for estimating PPE is suggested and applied to a practical example. A method of obtaining upper and lower bounds for the PPE is suggested.<<ETX>>

Victor L. Brailovsky | V. L. Brailovsky

[1] J. W. Gorman,et al. Selection of Variables for Fitting Equations to Data , 1966 .

[2] S. Fienberg,et al. Efficient Calculation of All Possible Regressions , 1968 .

[3] G. Diehr,et al. Approximating the Distribution of the Sample R2 in Best Subset Regressions , 1974 .

[4] Robert G. Miller. Statistical prediction by discriminant analysis , 1962 .

[5] A. C. Rencher,et al. Inflation of R2 in Best Subset Regression , 1980 .

[6] Murray Aitkin,et al. Simultaneous Inference and the Choice of Variable Subsets in Multiple Regression , 1974 .

[7] G. M. Furnival. All Possible Regressions with Less Computation , 1971 .

[8] V. L. Brailovsky. A Predictive Probabilistic Estimate for Selecting Subsets of Regressor Variables , 1987 .