Kernel regression is a non-parametric technique that is useful when the distribution of the observed data is unknown. In this project, I wrote two univariate kernel regression models: Gaussian Kernel and Epanechnikov Kernel. The two models were compared to a multi-dimensional Kernel in terms of prediction accuracy. Comparison of Gaussian and Epanechnikov Kernel Models Link to project 2

Generalized AutoRegressive Conditional Heteroskedasticity (GARCH) models are good for times series data that are very volatile. In this project, I developed and made predictions with different GARCH models using Apple stock prices. Time series chart Link to project 5