Using Gradient Boosting Machines, Bagging, Random Forest, and Bayesian Additive Regression Trees to predict wins in League of Legends. 10-fold CV LASSO regression for variable selection.

Coding Inverse CDF, Rejection, and Gibbs sampling

Provided point estimates and 95% confidence interval for probability of having anastomotic leak following colorectal surgery. Variables which increase leak probability discussed.

Consult tech start up on spending on advertising and revenue. Confidence and Prediction intervals provided, consultancy was based on client's budget for advertisement. Discussion of client expectation of analysis in relation to the limitations of the provided dataset.

Implemented the EM algorithm in this context for a multivariate normal distribution to estimate the mean and covariance using trivariate normal data.

Analyzing missile data and evaluating difference of means confidence intervals using kernel regression, cross-validation, and non-parametric bootstrapping.