Hemant Ishwaran

Research Interests

Cancer staging, Forests, Bayesian and frequentist variable selection, High throughput genomic data, Mixture models, Nonparametric Bayes

Staff, Dept. of Quantitative Health Sciences, Cleveland Clinic

Adjunct Professor, Dept. of Statistics, Case Western Reserve University

External Activities

Associate Editor, JASA, Theory and Methods, 05-11
Associate Editor, Electronic Journal of Statistics, 07-13
Associate Editor, Statistics and Probability Letters, 07-10
Web Editor, Inst. of Mathematical Statistics, 03-05

Brief Biography

PhD Statistics, Yale University, 1993
MSc Applied Statistics, Oxford University, 1988
BSc Mathematical Statistics, Univ. of Toronto, 1987

Some Recent Papers

Ishwaran H., Kogalur U.B., Moore R.D., Gange S.J. and Lau B.M. (2010). Random survival forests for competing risks.

Ishwaran H., Kogalur U.B., Chen X. and Minn A.J. (2010). Random survival forests for high-dimensional data. Submitted.

Ishwaran H. and Rao J.S. (2010). Generalized ridge regression: geometry and computational solutions when p is larger than n. Submitted.

Ishwaran H., Kogalur U.B., Gorodeski E.Z., Minn A.J. and Lauer M.S. (2010). High-dimensional variable selection for survival data. J. Amer. Stat. Assoc, 105, 205-217. pdf

Ishwaran H., Blackstone E.H., Hansen. C.A. and Rice T.W. (2009). A novel approach to cancer staging: application to esophageal cancer. Biostatistics, 10, 603-620. pdf

Ishwaran H., James L.F. and Zarepour M. (2009). An alternative to the m out of n bootstrap. J. Stat. Plann. Inference, 139, 788-801. pdf

Ishwaran H., Kogalur U.B., Blackstone E.H. and Lauer M.S. (2008). Random survival forests. Ann. Appl. Statist., 2, 841-860. pdf

Ishwaran H. and Rao J.S. (2005). Spike and slab variable selection: frequentist and Bayesian strategies. Ann. of Stat., 33, 730-773. pdf

BAMarray (3.0)

Java software for microarray data. Implements Bayesian Analysis of Variance for Microarrays (BAM)

randomSurvivalForest

Random survival forests R package for right-censored and competing risks data (release 3.6.3). pdf

3.0.x : Adaptive tree data imputation.

3.2.x: Joint VIMP (variable importance).

3.5.x : Factors with unlimited levels. Fast random splitting rules. 64-bit R.

3.6.x: Competing risks. Automatic variable selection; includes p bigger than n scenario. GUI for visualizing forests.

randomForestSRC

Random forests R package for survival, regression and classification. Coming soon!


spikeslab

Spike and slab R package for high-dimensional linear regression models (release 1.1.1). Uses a generalized elastic net for variable selection. Parallel processing enabled.