Stuart G Baker, ScD

Stuart G Baker, ScD

Portrait of Stuart G Baker, ScD
Stuart G Baker, ScD
Mathematical Statistician
(240) 276-7147
(240) 276-7845
5E606

View publications by Stuart G Baker

Research Highlights

Topic area Description
Bioinformatics Baker (2014, Microarrays) [Software #1] formulated a comparative analysis of biologically relevant response curves in gene expression experiments.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5003454/pdf/microarrays-03-00039.pdf
Cancer screening evaluation Baker et al. (2002, BMC Med Res Methodol) provided statistical guidance for the analysis of cancer screening data. Current work with Phil Prorok involves estimating the probability of overdiagnosis and ascertaining bias [Software #3].
Causal inference Baker and Lindeman (1994, Stat Med) developed the paired availability design. Swanson et al (2018 JASA) called this paper "seminal" along with Imbens and Angrist (1994) and Angrist, Imbens, and Rubin (1996).
Carcinogenesis Baker (2018 Organisms. J Bio Sciences) proposed the detached pericyte hypothesis, a novel explanation for many puzzling phenomena in tumorigenesis.
https://ojs.uniroma1.it/index.php/Organisms/article/view/14273/14040
Categorical data analysis Baker (1994, JRSS-D) developed the Multinomial-Poisson transformation to simplify computations.
Graphical methods Baker and Kramer (2001, J Women's Health Gend Based Med) formulated what was later called BK-Plot, which initiated later extensions.
Markers for early detection of cancer Baker (2003, BMC Med Res Meth) discussed the design and analysis of early detection studies.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC100327/pdf/1471-2288-2-4.pdf
Current work involves a method to substantially reduce sample size.
Markers for risk prediction Baker et al (2009, JRSS-A) developed relative utility curves and test tradeoffs leading to Baker (2017, Med Dec Making) [Software #4].
Markers for treatment selection Baker and Bonetti (2016, JNCI) developed subgroup methods for randomized trials based on decision-analysis. [Software #5, Software #6].
Missing data methods Baker and Laird (1988, JASA) developed the method to adjust for non-ignorable missing binary outcomes. Baker developed composite linear models for analyzing partially observed categorical data [Software #2]. Baker et al (2006, Biostatistics) developed the propensity-to-be-missing score for randomized trials.
Surrogate endpoints Baker (2017, Stat Med) developed 5 criteria for surrogate endpoint evaluation [Software #8].
Survival analysis Baker et al (1993, Biometrics) developed methodology for informative censoring and double sampling.
Twin studies Baker el al (2005, 2016, Biometrics) formulated the first fundamentally new twin method since R.A. Fisher developed variance components in the early 1900's. [Software #7].
Value of information Baker and Heidenberger (1988, Med Dec Making) developed a comprehensive value of information approach for planning randomized trials.

Awards and Honors

Dr. Baker was the first recipient of the distinguished alum award from the Department of Biostatistics at the Harvard School of Public Health. He is also a fellow of the American Statistical Association and an elected member of the International Statistical Institute.

Mathematica Packages

  1. Comparative Evaluation of Two Serial Gene Expression Experiments
  2. Composite Linear Models
  3. Estimating the Overdiagnosis Fraction in Cancer Screening
  4. Evaluating Risk Prediction Markers via Relative Utility Curves
  5. Evaluating Predictive Markers in a Randomized Trial with Binary Outcomes
  6. Evaluating Predictive Markers in a Randomized Trial with Survival Outcomes
  7. The Latent Class Twin Method
  8. Predicting Treatment Effect from Surrogate Endpoints and Historical Trials
  9. Simple and Flexible Classification of Gene Expression Microarrays Via Swirls and Ripples
  10. The Paired Availability Design and Related Instrumental Variable Meta-analyses