Michael Karcher
Education
- Ph.D., Statistics, University of Washington
- B.A., Mathematics, Swarthmore College
Teaching Interests
I teach because I loved the statistics courses I’ve taken, and I want to share that with new generations of students. In my undergraduate studies, I loved probability and mathematical statistics, and so I am excited to get to teach those classes here. In graduate school, I enjoyed stochastic modeling enough to build my dissertation on it, and one of the things that drew me to Muhlenberg was its exciting new graduate program with opportunities to teach about stochastic modeling.
I must also acknowledge that, unfortunately, not every student has such a positive relationship with statistics. Introductory statistics in particular has a somewhat harsh reputation as a difficult class, which I believe is not inherently warranted. In my teaching, I seek to make all of my statistics classes clear and accessible to all my students in order to share the enjoyment I had learning it.
Research and Scholarship
Most of my research has been in population genetics and phylodynamics. Both fields are related by using genetic sequence data samples (DNA or RNA from animals or viruses, for instance) to reconstruct a kind of family tree, called a genealogy, interrelating the ancestries of the members of the population that the samples were drawn from. You then have the option to use the genealogy to calculate estimates of the population size over time.
In particular, I built a method for detecting if the samples were drawn using convenience sampling (drawn proportionally to the population) and if so, incorporating the extra information into the analysis. In more recent work, I developed a method for combining genealogies on overlapping samples into a set of genealogies on all of the samples, often referred to as supertrees.
I have also spent some time doing research in cosmology and astrostatistics. I helped scientists design simulations of the early universe, tuning their models so as to end up matching as closely as possible the universe we observe today.
- Special Topic: Data Wrangling
- Statistical Analysis
- Statistical Models
- Topics in Advanced Modeling
Karcher, M. D., Carvalho, L. M., Suchard, M. A., Dudas, G., & Minin, V. N. (2020). Estimating effective population size changes from preferentially sampled genetic sequences. PLoS computational biology, 16(10), e1007774.
Karcher, M. D. (2018). Preferential sampling and model checking in phylodynamic inference (Doctoral dissertation).
Tremmel, M, Karcher, M., Governato, F, Volonteri, M, Quinn, T, Pontzen, A, Anderson, L (2017). The Romulus Cosmological Simulations: A Physical Approach to the Formation, Dynamics and Accretion Models of SMBHs. Monthly Notices of the Royal Astronomical Society, 470(1).
Anderson, L, Governato, F, Karcher, M, Quinn, T, Wadsley, J (2017). The Little Galaxies that Could (Reionize the Universe): Predicting Faint End Slopes & Escape Fractions at z > 4. Monthly Notices of the Royal Astronomical Society, 468(4).
Karcher, M, Palacios JA, Bedford T, Suchard MA, Minin VN (2016). Quantifying and Mitigating the Effect of Preferential Sampling on Phylodynamic Inference. PLoS Computational Biology 12(3): e1004789.
Karcher, M, Palacios JA, Lan, S, Minin VN (2016). PHYLODYN: an R package for phylodynamic simulation and inference. Molecular Ecology Resources 17(1): 96-100.
Lan, S, Palacios, JA, Karcher, M, Minin, VN, and Shahbaba, B (2015). An efficient Bayesian inference framework for coalescent-based nonparametric phylodynamics. Bioinformatics 31(20): 3282-3289.
Data Analytics
Mathematics, Computer Science & Statistics
Statistics
Contact: michaelkarcher@muhlenberg.edu