This Project is was created to compare a Linear Regression model from WEKA with XGBoost. XGBoost regression does poorly on time - series based data, thus we approached the solution by creating a n-step ahead algorithm to predict every y label.
Components invovled are: - Normalization - PCA - N Step ahead data processing - N Step ahead model creation - XGBoost model initialization based on docs