Ernest Fokoue Headshot

Ernest Fokoue

Associate Professor
School of Mathematical Sciences
College of Science

Office Location

Ernest Fokoue

Associate Professor
School of Mathematical Sciences
College of Science


I have been around Mathematical Sciences from my very early childhood, and I have always greatly enjoyed the beauty and power that mathematics affords to problem solving. Statistical Science in particular and specifically Statistical Machine Learning and Data Science are my passion, my profession and to some extent my vocation. The flexibility and power of probabilistic reasoning along with the richness and vast applicability of statistical models both fascinate and enthuse me immensely. My mission as a professor of statistical science is to generously, joyfully and rigorously share the full extent of my knowledge and expertise in this exciting field with the ultimate goal of empowering all the students that God graciously sends my way. Indeed, I always envision my students acquiring state-of-the-art knowledge in statistical science and ultimately developing into world class experts and pioneers.


Currently Teaching

3 Credits
The emphasis of this course is how to make valid statistical inference in situations when the typical parametric assumptions no longer hold, with an emphasis on applications. This includes certain analyses based on rank and/or ordinal data and resampling (bootstrapping) techniques. The course provides a review of hypothesis testing and confidence-interval construction. Topics based on ranks or ordinal data include: sign and Wilcoxon signed-rank tests, Mann-Whitney and Friedman tests, runs tests, chi-square tests, rank correlation, rank order tests, Kolmogorov-Smirnov statistics. Topics based on bootstrapping include: estimating bias and variability, confidence interval methods and tests of hypothesis.
3 Credits
This course introduces the students to the fundamental principles of modern graduate level statistical theory with a strong emphasis on conceptual aspects of estimation theory and statistical inference along with an exploration of the modern computational techniques needed in the application/implementation of the methods covered. Topics include fundamentals of probability theory for statistics, random variable with a focus on the understanding and use of probability distribution function (both probability density function and cumulative distribution function), quantiles of a distribution, understanding and use of the mathematical expectation operator, special discrete and continuous distributions, and distributions of functions of random variables and their use in statistical modelling, sums of random variables as used in statistics, point estimation, limit theorems, properties of estimators (bias, variance, mean squared error, consistency, efficiency, sufficiency), bias variance trade-off, interval estimation, hypothesis testing, bootstrap approach to estimation and inference, and elements of computational statistics.
3 Credits
This course is designed to provide the student with a solid practical hands-on introduction to the fundamentals of time series analysis and forecasting. Topics include stationarity, filtering, differencing, time series decomposition, time series regression, exponential smoothing, and Box-Jenkins techniques. Within each of these we will discuss seasonal and nonseasonal models.
0 - 9 Credits
Masters-level research by the candidate on an appropriate topic as arranged between the candidate and the research advisor.
3 Credits
A course that studies how a response variable is related to a set of predictor variables. Regression techniques provide a foundation for the analysis of observational data and provide insight into the analysis of data from designed experiments. Topics include happenstance data versus designed experiments, simple linear regression, the matrix approach to simple and multiple linear regression, analysis of residuals, transformations, weighted least squares, polynomial models, influence diagnostics, dummy variables, selection of best linear models, nonlinear estimation, and model building.
3 Credits
This course covers topics such as clustering, classification and regression trees, multiple linear regression under various conditions, logistic regression, PCA and kernel PCA, model-based clustering via mixture of gaussians, spectral clustering, text mining, neural networks, support vector machines, multidimensional scaling, variable selection, model selection, k-means clustering, k-nearest neighbors classifiers, statistical tools for modern machine learning and data mining, naïve Bayes classifiers, variance reduction methods (bagging) and ensemble methods for predictive optimality.

Latest News

Select Scholarship

Journal Paper
Corsetti, Matthew and Ernest Fokoue. "Nonnegative Matrix Factorization With Zellner Penalty." Open Journal of Statistics 5. 7 (2015): 777-786. Web. * £
Liu, Bohan and Ernest Fokoue. "Random Subspace Learning Approach to High-Dimensional Outliers Detection." Open Journal of Statistics 5. 6 (2015): 618-630. Web. * £
Yu, Xingchen and Ernest Fokoue. "Probit Normal Correlated Topic Model." Open Journal of Statistics 4. 11 (2014): 879-888. Web. * £
Fokoue, Ernest. "A Taxonomy Of Big Data For Optimal Predictive Machine Learning And Data Mining." Serdica Journal of Computing. (2014): 111-136. Print. * ∆ £
Bill, Jo and Ernest Fokoue. "A Comparative Analysis Of Predictive Learning Algorithms On High-Dimensional Microarray Cancer Data." Serdica Journal of Computing. (2014): 137-168. Print. * ∆ £
Ma, Zichen and Ernest Fokoue. "Accent Recognition for Noisy Audio Signals." Serdica Journal of Computing. (2014): 169-182. Print. * ∆ £
Ma, Zichen and Ernest Fokoue. "A Comparison Of Classifiers In Performing Speaker Accent Recognition Using MFCCs." Open Journal of Statistics 4. 4 (2014): 258-266. Web. * £
Fokoue, Ernest. "Stable Radial Basis Function Selection via Mixture Modelling of the Sample Path." Journal of Data Science 9. 3 (2011): 345-358. Web. * £
Fokoue, Ernest, Dongchu Sun, and Prem Goel. "Fully Bayesian Analysis of the Relevance Vector Machine With Extended Prior." Statistical Methodology 8. (2011): 83-96. Print. * £
Fokoue, Ernest and Prem Goel. "An Optimal Experimental Design Perspective on Radial Basis Function Regression." Communications in Statistics: Theory and Methods 40. 6 (2011): 1-12. Print. « £
Fokoue, Ernest and Bertrand Clarke. "Bias-Variance Trade-off for Prequential Model List Selection." Statistical Papers 52. 4 (2011): 813-833. Print. * £
Fokoue, Ernest. "Beta Induced Sparsity Algorithm." Advanced and Applications in Statistical Science. (2011): 1-4. Web. * £
Book Chapter
Dey, Tanujit and Ernest Fokoue. "“Bayesian Variable Selection For Predictively Optimal Regression." Current Trends in Bayesian Methodology with Applications. Ed. Dipak K, Umesh Singh and A. Loganathan. New York, New York: Chapman and Hall, 2015. Print. * ∆ £