Dr. Ayesha Ali received her PhD in Statistics from the University of Washington in 2002, after which she became an Assistant Professor in the Department of Statistics and Applied Probability at the National University of Singapore. Ali joined the University of Guelph’s Department of Mathematics and Statistics in 2006, where she is now an Associate Professor of Statistics.
Ali studies complex systems for which the challenge lies in inferring links in a high dimensional network based on limited observed data. She approaches this problem by first considering the associated data generating process and relating it to a graphical Markov model. Her research typically involves strong statistical computing skills. Specific applications include:
Link formation in plant-pollinator networks. Ali has developed a grouped Dirichlet-multinomial regression model with regularization for model selection and adapted latent Dirichlet allocation to the plant-pollinator network setting. Future work involves accommodating sampling weights, zero-inflation, and the development of statistical methods to compare networks in space and/or time.
Model selection for animal health. Ali is interested in several applied problems for the genetic selection of livestock. Her research team has developed models that exploit the graphical structure of SNPs, lipidomics, or mid-infrared spectral data for the prediction of an outcome, whether continuous or binary, using regularization. A specific application involves the selection of candidate genes in the prediction of boar taint. Her group has also analyzed milk spectral data for the purpose of improving the nutraceutical properties of milk and overall health of cows. The project also involves developing methods to link the spectral data with genetic, physical and performance traits in a longitudinal setting.
Bootstrapping in longitudinal studies. Causal models can be used to provide unbiased parameter estimates in the presence of time-dependent confounding. Ali has applied marginal structural models to natural history data and developed a Cox score bootstrap for fast, reliable bootstrapping to be used in large scale simulation studies. She is also developing bootstrap-based methods for standard errors in cure rate models and confidence intervals for cure fractions in first hitting time models.
The Canadian Journal of Statistics Award, 2020 for “Doubly sparse regression incorporating graphical structure among predictors.”
NSERC Discovery Grant, 2018
NSERC Collaborative Research and Development Grant, 2015
Senior Program Committee member, Uncertainty in Artificial Intelligence Conference, 2015-2017
Publications Officer and Newsletter Editor, The International Environmetrics Society, 2011-2017