Elif Acar
Associate Professor, Statistics; Canada Research Chair in One Health Approaches to Statistical Modelling
College of Computational, Mathematical and Physical Sciences, Department of Mathematics & Statistics
Research Themes
-
High-dimensional dependence modelling
Many problems in data science involve multivariate data of a complex nature and in massive amounts. While the advancements in computing and storage technologies enable automated statistical analysis of large amounts of data, these investigations are often limited to univariate features, lacking an understanding of multivariate dependencies among outcomes of interest. Dr. Acar’s research involves development of multivariate modelling strategies that can successfully capture and account for statistical dependence under various scenarios of data complexity. Specific interests include flexible modelling of incomplete data types and automated methods of multivariate analysis.
-
Research Synthesis and Replicability in High-Throughput Studies
With the increasing shift towards consortium research in many domains, there has been a growing interest in methods for combining and synthesizing statistical evidence from multiple studies. This is typically achieved using meta-analysis. Dr. Acar’s research involves development of meta-analysis methods for synthesizing statistical evidence in high-throughput studies in three interrelated domains: genome-wide association studies, microbiome studies and high-throughput phenotyping experiments.
Opportunities: Dr. Acar is accepting postdoctoral fellows, PhD and MSc students. Interested candidates should contact Dr. Acar with their CV and cover letter.
Interests
- High-dimensional dependence
- Statistical genetics
- Nonparametric estimation
- Survival analysis
- Longitudinal data analysis
- Meta-analysis
- Replicability
Education
- PhD, Statistics, University of Toronto, 2010
- MSc, Mathematics, University of New Hampshire, 2006
- BSc, Statistics, Middle East Technical University, 2003
Recent Publications
- Hoque M E, Acar E, Torabi M (2023). A time heterogeneous D-vine copula model for unbalanced and unequally spaced longitudinal data. Biometrics, 79(2), 734–746. doi:10.1111/biom.13652
- Loureiro-Rodríguez V, Acar E (2022). The Matched-Guise Technique. Research Methods in Language Attitudes edited by Ruth Kircher and Lena Zipp, Cambridge University Press. doi:10.1017/9781108867788.016
- Enikanolaiye A, Ruston J, Zeng R, Taylor C, Shrock M, Buchovecky C M, Shendure J, Acar E, Justice M J (2020). Suppressor mutations in Mecp2-null mice reveal that the DNA damage response is key to Rett syndrome pathology. Genome Research, 30(4), 540—552. doi:10.1101/gr.258400.119
- Grover K, Acar E, Torabi M (2020). Copula-based predictions in small area estimation. Canadian Journal of Statistics, 48(4), 685–711. doi:10.1002/cjs.11558
- Acar E, Czado C, Lysy M (2019). Flexible dynamic vine copula models for multivariate time series data. Econometrics and Statistics, 12, 181–197. doi:10.1016/j.ecosta.2019.03.002
- Acar E, Azimaee P, Hoque M E (2019). Predictive assessment of copula models. Canadian Journal of Statistics, 47(1), 8–26. doi:10.1002/cjs.11468
Awards
- MATRIX-Simons Travel Award, 2024.
- NSERC Discovery Grant, 2020-2027.
- Hasselt University Travel Award, 2015.
- Women for Math Science Award, Technische Universität München, 2014.
- CANSSI Collaborative Research Team Project Grant, 2014-2017.
- NSERC Discovery Grant with Early Career Supplement, 2013-2020.
- David P. Byar Travel Award, American Statistical Association, 2010.