Annals of Statistics

“Preconditioning” for feature selection and regression in high-dimensional problems

Debashis Paul, Eric Bair, Trevor Hastie, and Robert Tibshirani

Full-text: Open access


We consider regression problems where the number of predictors greatly exceeds the number of observations. We propose a method for variable selection that first estimates the regression function, yielding a “preconditioned” response variable. The primary method used for this initial regression is supervised principal components. Then we apply a standard procedure such as forward stepwise selection or the LASSO to the preconditioned response variable. In a number of simulated and real data examples, this two-step procedure outperforms forward stepwise selection or the usual LASSO (applied directly to the raw outcome). We also show that under a certain Gaussian latent variable model, application of the LASSO to the preconditioned response variable is consistent as the number of predictors and observations increases. Moreover, when the observational noise is rather large, the suggested procedure can give a more accurate estimate than LASSO. We illustrate our method on some real problems, including survival analysis with microarray data.

Article information

Ann. Statist., Volume 36, Number 4 (2008), 1595-1618.

First available in Project Euclid: 16 July 2008

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62J07: Ridge regression; shrinkage estimators

Model selection prediction error lasso


Paul, Debashis; Bair, Eric; Hastie, Trevor; Tibshirani, Robert. “Preconditioning” for feature selection and regression in high-dimensional problems. Ann. Statist. 36 (2008), no. 4, 1595--1618. doi:10.1214/009053607000000578.

Export citation


  • [1] Bair, E., Hastie, T., Paul, D. and Tibshirani, R. (2006). Prediction by supervised principal components. J. Amer. Statist. Assoc. 101 119–137.
  • [2] Bair, E. and Tibshirani, R. (2004). Semi-supervised methods to predict patient survival from gene expression data. PLOS Biology 2 511–522.
  • [3] Donoho, D. (2004). For most large underdetermined systems of equations, the minimal 1-norm solution is the sparsest solution. Technical report, Stanford Univ.
  • [4] Donoho, D. and Elad, M. (2003). Optimally sparse representation from overcomplete dictionaries via 1-norm minimization. Proc. Natl. Acad. Sci. USA 100 2197–2202.
  • [5] Efron, B., Hastie, T., Johnstone, I. and Tibshirani, R. (2004). Least angle regression (with discussion). Ann. Statist. 32 407–499.
  • [6] Fan, J. and Li, R. (2005). Variable selection via nonconcave penalized likelihood and its oracle properties. J. Amer. Statist. Assoc. 96 1348–1360.
  • [7] Fan, J. and Peng, H. (2004). Nonconcave penalized likelihood with a diverging number of parameters. Ann. Statist. 32 928–961.
  • [8] Kalbfleisch, J. and Prentice, R. (1980). The Statistical Analysis of Failure Time Data. Wiley, New York.
  • [9] Knight, K. and Fu, W. (2000). Asymptotics for lasso-type estimators. Ann. Statist. 28 1356–1378.
  • [10] Meinshausen, M. (2005). Lasso with relaxation. Research Report 129, ETH Zürich.
  • [11] Meinshausen, N. and Bühlmann, P. (2006). High-dimensional graphs and variable selection with the lasso. Ann. Statist. 34 1436–1462.
  • [12] Osborne, M., Presnell, B. and Turlach, B. (2000). On the lasso and its dual. J. Comput. Graph. Statist. 9 319–337.
  • [13] Park, M. Y. and Hastie, T. (2006). An l1 regularization-path algorithm for generalized linear models. Unpublished manuscript.
  • [14] Paul, D. (2005). Nonparametric estimation of principal components. Ph.D. dissertation, Dept. Statistics, Stanford Univ.
  • [15] Shen, X. and Ye, J. (2002). Adaptive model selection. J. Amer. Statist. Assoc. 97 210–221.
  • [16] Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. J. Roy. Statist. Soc. Ser. B 58 267–288.
  • [17] Tibshirani, R., Hastie, T., Narasimhan, B. and Chu, G. (2001). Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc. Natl. Acad. Sci. USA 99 6567–6572.
  • [18] Zhao, H., Tibshirani, R. and Brooks, J. (2005). Gene expression profiling predicts survival in conventional renal cell carcinoma. PloS. Med. 3(1) e13.
  • [19] Zhao, P. and Yu, B. (2006). On model selection consistency of Lasso. J. Mach. Learn. Res. 7 2541–2563.
  • [20] Zou, H. (2005). The adaptive lasso and its oracle properties. J. Amer. Statist. Assoc. 101 1418–1429.