Stewart J Anderson

  • Professor Emeritus

Areas of current methodological research interests include: 1) methods in clinical trials including survival analysis, Bayesian clinical trials design, and the use of Markov processes in modeling clinical data; 2) modern regression techniques especially nonparametric regression and regularization methods; and 3) general methodology in longitudinal data analysis.

I am also involved in several large scale collaborative projects. I was the primary statistician for many breast cancer treatment protocols for the National Surgical Adjuvant Breast and Bowel Project (NSABP) . More than 15,000 patients were followed across these trials. I am now an Associate Director of the Division of Biostatistics and Science for the newly formed cancer cooperative group, NRG Oncology. In addition to my involvement in cancer research, Until recently, I had been a co-director of the Research Design and Biostatistics Unit for the ACISR, a large center for prevention of adult late-life depression, headed by Dr. Charles Reynolds in the Department of Psychiatry in the University of Pittsburgh School of Medicine. In this capacity, I oversaw the design, analysis and development of methodology for three prospective randomized studies that compare strategies for preventing depression in older adults. Currently, I am involved with a group that investigates the aging brain with respect to amyloid deposits and other features of brain scans. Some of this work involves developing statistical methodology to distinguish a small number of important features in brain imagining datasets that often consist of extremely large numbers of features that potentially could be related to aging.


1978 | University of Colorado, Boulder CO | Bachelor of Arts, Mathematics
1981 | University of Kansas, Lawrence KS | Master of Arts, Mathematics
1987 | University of Colorado Health Sciences Center, Denver CO | Doctor of Philosophy, Biometrics


BIOST 241, 2041 Introduction to Statistical Methods 1
A survey course in applied statistics taught to first year graduate students in biostatistics, public health and to postgraduate medical fellows. Covered descriptive statistics, probability (axioms, counting, conditional probability, Bayes' theorem, normal, uniform, binomial and Poisson distributions), estimation and confidence intervals, one sample and two sample hypothesis testing for normally distributed data and for proportions, general introduction to one-way ANOVA, contingency tables, simple linear regression, correlation, and statistical packages (Minitab, SAS, R). Texts used: Fisher and van Belle Biostatistics: A Methodology for the Health Sciences, 1993. Rosner, Fundamentals of Biostatistics, 5th ed., 1999. Also, extensive class notes were distributed. I taught this class every fall from 1989 to 1997 and then again in summer 2006.

BIOST 242, 2042 Introduction to Statistical Methods 2
The second part of the sequence in biostatistics (complements BIOST 2041). Covered nonparametric methods, multiple linear regression, odds ratio, relative risk, logistic regression, Mantel-Haenszel procedure, methods in survival analysis (Kaplan-Meier, life table analyses), multiple comparisons procedures, general linear contrast, multi-way ANOVA (general fixed effects model, factorial design, interactions, randomized blocks, random effects models, repeated measures) and analysis of covariance. Texts used: Fisher and van Belle Biostatistics: A Methodology for the Health Sciences, 1993; Rosner, Fundamental of Biostatistics, 5th ed., 1999. Also, extensive class notes distributed via the internet. From these notes, I published a text book; Anderson, SJ. Biostatistics: A Computing Approach, Taylor & Francis Group, LLC, Boca Raton, FL, 2011. ISBN 978-1-58488-834-5.. I taught this course every spring from 1990 to 2002 and again, in Spring 2015. I will teach this again in spring 2016.

BIOST 2052 Multivariate Analysis
This course covers advanced topics in multivariate analysis including the multivariate normal distribution, estimation of the mean vector and the covariance matrix, distributions and uses of simple, partial and multiple correlation coefficients, the generalized T2 statistic, the distribution of the sample generalized variance, multivariate analysis of variance and applications of multivariate methods to repeated measures analysis, factor analysis or discriminant analysis. The emphasis in the beginning of the course is on theory. Later, applications and computational methods are emphasized. Students are required to prepare a project on an application area which requires them to summarize and reproduce results from the literature. Texts used or referred to in the course include Anderson TW. An Introduction to Multivariate Statistical Analysis, 3rd edition, 2003; Mardia, K.V., Kent, JJ and Bibby, JM. Multivariate Analysis,Academic Press, 1979; Khattree, R. and Naik, D.N. Applied Multivariate Statistics with SAS Software, 2nd edition, SAS Institute, Inc, Cary, NC, 1999; and Searle, S.R. Matrix Algebra Useful for Statistics, Wiley & Sons, 1982. In addition to the texts, extensive class notes are distributed by the instructor. I taught this course roughly every other fall from 1999 to 2012.

BIOST 2096 (formerly BIOST 2075) Numerical Methods in Biostatistics
The purpose of this course is to familiarize the student with a wide range of numerical methods which are useful in biostatistical research. Selected computational techniques used in statistical research will be covered in the course. Some background will be provided to facilitate understanding of a few numerical algorithms widely used in statistics. The following topics are covered: recurrence relations, power series and asymptotic expansions, generating pseudo-random deviates, basic simulation methodology, use of antithetic variables, importance sampling, solutions of nonlinear equations, Newton’s method, vector and matrix norms, linear regression and matrix inversion, other topics in multivariate analysis, splines, finite-state Markov chains, Hastings-Metropolis algorithms and Gibbs sampling. The goal of this course is for the student to gain knowledge useful in understanding and developing statistical methodology. Upon completion of this course, the student should have a clear understanding of how to conceptualize and implement computer simulations useful for doing statistical or other quantitative research. The primary texts used for this course are Lange, Numerical Analysis for Statisticians, Second edition, Springer-Verlag, 2010 and Ross, Simulation, Fifth Edition, Academic Press, 2013. Extensive class notes are distributed. As prerequisites, students should have had one year of courses in mathematical statistics and some experience with at least one computer language (e.g., Fortran, C, Pascal or Java) and some knowledge of one or more of the following statistical packages: R, SAS, S-plus, Stata, Matlab, Mathematica. I've taught this course once every 2 years from 2003 to 2013.

Doctoral Students

  1. Joanne Beer, Ph.D. (co-advisor with Rob Krafty) – Graduated August 2018. Dissertation: “Predicting clinical variables from neuroimages with fused sparse group LASSO” (Winner, best dissertation)
  2. John Pleis, Ph.D. – Graduated August 2018. Dissertation: “Mixtures of discrete and continuous variables: considerations for dimension reduction”
  3. Abraham Apfel, Ph.D. –Graduated June 2017.  Dissertation: “Stability analysis of sparse K-means”
  4. Andrew Potter, Ph.D. – Graduated April 2017. Dissertation: “Multiscale multivariate functional principal component analysis with an application to multivariate longitudinal cardiac signals”
  5. Jia-Yuh (Lily) Chen, Ph.D. – Graduated June  2016.  Dissertation: “Joint modeling of bivariate longitudinal and survival data in spouse pairs”
  6. Beth Zamboni, Ph.D. – Graduated April 2015.  Dissertation: “Twisted survival: identifying surrogate endpoints for mortality using QTWIST and conditional disease free survival
  7. Xiaoxue Li, Ph.D. – Graduated December 2014.  Dissertation: “Time-varying coefficient gap time models for Ecological Momentary Assessment data”.
  8. Yun Ling, Ph.D. – Graduated April 2014.  Dissertation: “Influential observation detection for multilevel multivariate growth curve models”
  9. Ji-in Choi, Ph.D. – Graduated August 2012.  Dissertation: “Prediction in the joint modeling of mixed types of multivariate longitudinal outcomes and a time-to-event outcome"
  10. Yoko Tanaka, Ph.D. – Graduated December 2010.  Dissertation: "An adaptive two-stage dose-response design method for establishing proof of concept in drug development"
  11. Folefac Atem, Ph.D. – Graduated August 2010. Dissertation: “Rationale for choosing an explicit correlation structure in a multilevel analysis with bivariate outcomes”
  12. Xing Yuan, Ph.D. – Graduated December 2009.  Dissertation: “A meta-analytic framework for combining incomparable Cox proportional hazard models caused by omitting important covariates”. Winner best dissertation in Department of Biostatistics
  13. Meredith (Lotz) Wallace, Ph.D. – Graduated August 2009.  Dissertation: “Modeling missing covariate data and temporal features of time-dependent covariates in tree-structured survival analysis”
  14. Fiona Callaghan, Ph.D.  (Co-chaired with Joyce Chang) – Graduated April 2008.  Dissertation: “Classification trees for survival data with competing risks"
  15. L. Scott Dean, Ph.D. – Graduated April 2007.  Dissertation: “A method for detecting optimal splits over time in survival analysis using tree-structured models”
  16. Jia Li, Ph.D. – Graduated August 2006.  Dissertation: “A strategy for stepwise regression procedures in survival analysis with missing covariates”
  17. Feng-shou Ko, Ph.D. – Graduated August 2006.  Dissertation: “Identification and assessment of longitudinal biomarkers using frailty models in survival analysis”
  18. Michael B. (Brent) McHenry, Ph.D. – Graduated June 2004.  Dissertation: “New estimation approaches in survival analysis with Aalen’s additive risk model”.  Winner of two ASA student travel awards.
  19. Mary Kelley, Ph.D. – Graduated December 2003.  Dissertation: “Zero inflation in ordinal data: applications of a mixture model”
  20. Maria K. Mor, Ph.D. – Graduated April 2003.   Dissertation: “A Bayesian group sequential approach for multiple endpoints”
  21. Yanming Yin, Ph.D. – Graduated April 2002.  Dissertation: “Tree-structured modeling for interval-censored survival data”
  22. Wei Tian, Ph.D. - Graduated June, 1999. Dissertation: “Markov chain models for analyzing multivariate repeated categorical data with incomplete observations
  23. Sang Ahnn, Ph.D. – Graduated December 1994. Dissertation: "Sample size determination for comparing more than two survival distributions". 
  24. Lingshi Tan, Ph.D. – Graduated April 1993. Dissertation: "A multivariate growth curve model with random effects and CAR(1) errors", winner of a 1994 Biometrics Society (ENAR) student travel award
Selected Publications


Anderson, SJ. Biostatistics: A Computing Approach, Taylor & Francis Group, LLC, Boca Raton, FL, 2011. ISBN 978-1-58488-834-5.

Selected Refereed Articles
Anderson SJ, Jones, RH, and Swanson, GD. Smoothing polynomial splines for bivariate data. SIAMJournal of Scientific and Statistical Computing 11(4), 1990.

Fisher B, Anderson S, Fisher E, Redmond C, et al. The significance of breast tumor recurrence following lumpectomy for the treatment of breast cancer: findings from NSABP B-06.The Lancet, 338, 327-331, 1991.

Anderson SJ and Jones, RH. Smoothing splines for longitudinal data. Statistics in Medicine, 14, 1235-1248, 1995.

Fisher B, Anderson S, Redmond C, Wolmark N, Wickerham DL, and Cronin W. Reanalysis and results after twelve years of follow-up in of a randomized clinical trial comparing total mastectomy to lumpectomy with and without irradiation in the treatment of breast cancer. The New England Journal of Medicine, 333, 1456-1461, 1995.

Ahnn S and Anderson S.&nbsp Sample size determination in complex clinical trials comparing more than two groups for survival endpoints. Statistics in Medicine, 17, 2525-2534, 1998.

Tian W and Anderson SJ. Markov chain models for multivariate repeated binary data analysis.Communications in Statistics, Simulation and Computation, 29(4), 2000.

Fisher B, Anderson S, Bryant J, Margolese RG, Deutsch J, Fisher ER, Jeong J-H and Wolmark N. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy followed by irradiation for the treatment of invasive breast cancer.The New England Journal of Medicine, 347(16), 1233-1241, 2002.

Mor MK and Anderson S. A Bayesian group sequential approach for multiple endpoints.Sequential Analysis, 24, 1-18, 2005.

Dang Q, Anderson S, Tan L and Mazumdar S. Modeling unequally spaced bivariate growth curve data using a Kalman filter approach.Communications in Statistics, Theory and Methods, 34(8), 1821-1831, 2005.

Anderson S. Some thoughts on the reporting of adverse events in phase II cancer clinical trials. Journal of Clinical Oncology, 24(24), 3821-3822, August 20, 2006. PMID # 16921032

Dang Q, Mazumdar S, Anderson SJ, Houck PR and Reynolds CF. Using trajectories from a bivariate growth curve as predictors in a Cox regression model.Statistics in Medicine, 26, 800-811, 2007. PMID 16612837

Kelley, M and Anderson, S. Zero inflation in ordinal data: incorporating susceptibility to response through the use of a mixture model. Statistics in Medicine, 27, 3674-3688, 2008. PMCID # 2572084

Anderson SJ, Wapnir I, Dignam JJ, Fisher,B, Mamounas,EP, Jeong J-H, Geyer,CE, Wickerham,DL, Costantino, JP, Wolmark, N. Prognosis after IBTR and locoregional recurrences in patients treated by breast conserving surgery in five NSABP node-negative protocols.Journal of Clinical Oncology 27(15), 2466-2473, 2009. PMID 19349544

Yuan X and Anderson S. Meta-analysis methodology for combining treatment effects from Cox proportional hazard models with different covariate adjustments. Biometrical Journal, 52(4), 519–537, 2010. DOI:10.1002/bimj.200900168.

Krag DN, Anderson SJ, Julian TB, Brown AM, Harlow SP, et al. Sentinel-lymph-node resection compared with conventional axillary-lymph-node dissection in clinically node-negative patients with breast cancer: overall survival findings from the NSABP B-32 randomized phase 3 trial.The Lancet Oncology, 11(10), 927-933, October 2010. PMID 20863759.

Reynolds CF III, Butters MA, Lopez O, Pollock BG, Dew MA, Mulsant BH, Lenze EJ, Holm M, Rogers JC, Mazumdar, S, Houck PR, Begley A, Anderson SJ, Karp JF, et al. maintenance treatment of depression in old age: a randomized, double-blind, placebo-controlled evaluation of the efficacy and safety of Donepezil combined with antidepressant pharmacotherapy.Archives of General Psychiatry, 68(1), 51–60, 2011. PMC3076045

Gildengers AG, Chisholm D, Anderson SJ, Begley, A, Holm M, Rogers J, Reynolds CF, Mulsant BH: Two year course of cognitive and IADL function in older adults with bipolar disorder: evidence for neuroprogression? Psychol Med. Jul 30:1-11, 2012, PMID 22846332

Wallace MEL, Anderson SJ, and Mazumdar S. Incorporating temporal features of repeatedly measured covariates into tree-structured survival models.Biometrical Journal, 54(2), 181–196, 2012.

Franchetti Y, Anderson SJ and Sampson AR. An adaptive two-stage dose-response design method for establishing proof of concept.J Biopharmaceutical Statistics, 23(5), 1124-1154, 2013. PMID 23957520. PMCID: 4073119

Mamounas EP, Anderson SJ, Dignam, JD§, Bear HD, Julian TB, Geyer CE, Jr. Taghian A, Wickerham DL, and Wolmark N. Predictors of loco-regional recurrence following neoadjuvant chemotherapy: results from combined analysis of NSABP B-18 and B-27. Journal of Clinical Oncology, 30(32), 3960 – 3966, Nov 10, 2012.

Shiffman S, Dunbar M, Kirchner T, Li X, Tindle H, Anderson S and Scholl S. Smoker reactivity to cues: effects on craving and on smoking behavior.Abnormal Psychology, 122(1), 264-280, Feb 2013. PMID: 22708884.

Aebi S, Gelber S, Anderson SJ, Láng I, Robidoux A, Martín M, Nortier JWR, Paterson AHG, Rimawi MF, Cañada JMB, Thürlimann B, Murray E, Mamounas EP, Geyer CE Jr, Price KN, Coates AS, Gelber RD, Rastogi P, Norman Wolmark N and Wapnir IL. Chemotherapy for isolated locoregional recurrence of breast cancer (CALOR): a randomised trial. Lancet Oncology, 15, 156–163, Feb 2014. PMID: 24439313.

Reynolds CF III, Thomas SB, Morse JQ, Anderson SJ, Albert SM, Dew MA, Begley AE, Karp JF, Gildengers A, Butters MA, Stack JA, Kasckow J, Miller MD, Quinn SC. Early Intervention to preempt major depression in older black and white adults. Psychiatric Services, 65, 765–773, 2014, 2014. Epub 2014 Mar 17. PMID: 24632760.

Choi J-I, Anderson SJ, Richards TJ and Thompson WK. Prediction of transplant-free survival in idiopathic pulmonary fibrosis patients using joint models for event times and mixed multivariate longitudinal data.Journal of Applied Statistics, 41 (10), 2192–2205, 2014. PMID:25214700 PMCID:PMC4157686.

Troy WC and Anderson SJ. Exploring the role of the host-tumor interactions in tumor growth and regression.Quarterly of Applied Mathematics, 73(1), 131-161, 2015.

Beer JC, Aizenstein HJ, Anderson SJ and Krafty RT.Incorporating prior information with fused sparse group LASSO: application to prediction of clinical measures from neuroimages. To appear in Biometrics.