Feature selection by higher criticism thresholding achieves the optimal phase diagram

Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences

Volumn 367, Issue 1906, 2009, Pages 4449-4470

  Donoho, David ^a   Jin, Jiashun ^b  

^a STANFORD UNIVERSITY   (United States)

^b CARNEGIE MELLON UNIVERSITY   (United States)

Author keywords

Asymptotic rare weak model; False discovery rate; Feature selection by thresholding; Fisher's separation measure; Linear classification; Phase diagram

Indexed keywords

ASYMPTOTIC ANALYSIS; DIELECTRIC DEVICES; PHASE DIAGRAMS; TWO DIMENSIONAL;

FALSE DISCOVERY RATE; FEATURE SELECTION; FISHER'S SEPARATION MEASURE; LINEAR CLASSIFICATION; THRESHOLDING;

PHASE SPACE METHODS;

EID: 73349141108     PISSN: 1364503X     EISSN: None     Source Type: Journal    
DOI: 10.1098/rsta.2009.0129     Document Type: Article

References (30)

1. Abramovich, F. & Benjamini, Y. 1995 Thresholding of wavelet coefficients as multiple hypotheses testing procedure. In Wavelets and statistics (eds A. Antoniadis & G. Oppenheim), pp. 5-14. New York, NY: Springer.
2. Abramovich, F., Benjamini, Y., Donoho, D. & Johnstone, I. 2006 Adapting to unknown sparsity by controlling the false discovery rate. Ann. Stat. 34, 584-653. (doi:10.1214/009053606000000074) (Pubitemid 44091387)
3. Anderson, T. W. 2003 An introduction to multivariate statistical analysis, 3rd edn. New York, NY: Wiley.
4. Benjamini, Y. & Hochberg, Y. 1995 Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. B 57,289-300.
5. Bickel, P. & Levina, E. 2004 Some theory of Fisher's linear discriminant function, 'naive Bayes', and some alternatives when there are many more variables than observations. Bernoulli 10,989-1010. (doi:10.3150/bj/ 110631487).
6. Cai, T., Jin, J. & Low, M. 2007 Estimation and confidence sets for sparse normal mixtures. Ann. Stat. 35,2421-2449. (doi:10.1214/ 009053607000000334).
7. Cayon, L., Jin, J. & Treaster, A. 2005 Higher criticism statistic: detecting and identifying non-Gaussianity in the WMAP first year data. Mon. Not. R. Astron. Soc. 362,826-832. (doi:10.1111/j.1365-2966.2005.09277.x). (Pubitemid 41337278)
8. Delaigle, A. & Hall, J. 2009 Higher criticism in the context of unknown distribution, non-independence and classification. In Perspectives in mathematical sciences I: Probability and statistics, Platinum Jubilee Proceedings of the Indian Statistical Institute (eds N. S. Narasimha Sastry, T. S. S. R. K. Rao, M. Delampady & B. Rajeev), Statistical Science and Interdisciplinary Research, vol.7,pp. 109-138. Hackensack, NJ: World Scientific.
9. Donoho, D. & Jin, J. 2004 Higher criticism for detecting sparse heterogeneous mixtures. Ann. Stat. 32, 962-994. (doi:10.1214/009053604000000265) . (Pubitemid 41250290)
10. Donoho, D. & Jin, J. 2006 Asymptotic minimaxity of false discovery rate thresholding for sparse exponential data. Ann. Stat. 34,2980-3018. (doi:10.1214/009053606000000920).
11. Donoho, D. & Jin, J. 2008 Higher criticism thresholding: optimal feature selection when useful features are rare and weak. Proc. Natl Acad. Sci. USA 105,14790-14795. (doi:10.1073/pnas. 0807471105).
12. Donoho, D. & Jin, J. In preparation. If useful features are rare and weak, HCT yields successful classification where success is possible.
13. q;-error. Prob. Theory Related Fields 2, 277-303.
14. Donoho, D.' Johnstone, I., Hoch, J. C. & Stern, A. S. 1992 Maximum entropy and the nearly black object. J. R. Stat. Soc. B 54,41-81.
15. Efron, B.' Tibshirani, R., Storey, J. & Tusher, V. 2001 Empirical Bayes analysis of a microarray experiment. J. Am. Stat. Assoc. 99,96-104. (doi:10.1198/016214504000000089).
16. Fan, J. & Fan, Y. 2008 High-dimensional classification using features annealed independence rules. Ann. Stat. 36,2605-2637. (doi:10.1214/07-AOS504).
17. Golub, T. et al. (1999) Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 286,531-536. (doi:10.1126/science.286. 5439.531).
18. Hall, P. & Jin, J. 2008 Properties of higher criticism under strong dependence. Ann. Stat. 36,381-402. (doi:10.1214/009053607000000767).
19. Hall, P. & Jin, J. 2009 Innovative higher criticism for detecting sparse signals in correlated noise. (http://arxiv.org/abs/0902.3837v1).
20. Hall, P., Pittelkow, Y. & Ghosh, M. 2008 Theoretical measures of relative performance of classifiers for high dimensional data with small sample sizes. J. R. Stat. Soc. B 70,158-173.
21. Hedenfalk, I. et al. 2001 Gene-expression profile in hereditary breast cancer. N. Engl. J. Med. 344,539-448. (doi:10.1056/NEJM200102223440801). (Pubitemid 32167867)
22. Ingster, Yu. I. 1997 Some problems of hypothesis testing leading to infinitely divisible distribution. Math. Methods Stat. 6,47-69.
23. Ingster, Yu. I., Pouet, C. & Tsybakov, A. B. 2009 Sparse classification boundaries. (http://arxiv.org/abs/0903.4807v).
24. Ingster, Yu. I., Pouet, C. & Tsybakov, A. B. 2009 Classification of sparse high-dimensional vectors. Phil. Trans. R. Soc. A 367,4427-4448. (doi:10.1098/rsta.2009.0156).
25. Jager, L. & Wellner, J. 2007 Goodness-of-fit tests via phi-divergences. Ann. Stat. 35,2018-2053.(doi:10.1214/0009053607000000244).
26. Jin, J. 2003 Detecting and estimating sparse mixtures. PhD thesis, Department of Statistics, Stanford university.
27. Jin, J. 2009 Impossibility of successful classification when useful features are rare and weak. Proc. Natl Acad. Sci. USA 106,8859-8864. (doi:10.1073/pnas.0903931106).
28. Jin, J., Starck, J.-L., Donoho, D., Aghanim, N. & Form, O. 2005 Cosmological non-Gaussian signature detection: comparing performances of different statistical tests. EURASIP J. Appl. Signal Process 15,2470-2485. (doi:10.1155/ASP.2005.2470). (Pubitemid 43153758)
29. Johnstone, I. & Silverman, B. 2004 Needles and straw in haystacks: empirical Bayes estimates of possibly sparse sequences. Ann. Stat. 32,1594-1649. (doi:10.1214/009053604000000030). (Pubitemid 41250278)
30. Tibshirani, R., Hastie, T., Narasimhan, B. & Chu, G. 2002 Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc. Natl Acad. Sci. USA 99, 6567-6572. (doi:10.1073/pnas.082099299). (Pubitemid 34526173)