Time Series Analysis and Forecasting of the US Housing Starts using Econometric and Machine Learning Models
In this thesis, I have performed time series analysis and forecasted the monthly value of housing starts for the year 2019 using several econometric and machine learning algorithms. In the rapidly emerging field of artificial intelligence, data scientists are heavily improvising and using machine learning models to predict any variable. I have applied a few popular techniques from machine learning to predict housing starts for the year 2019. Some of these methods are - artificial neural networks, ridge regression, K-Nearest Neighbors, and support vector regression, and created an ensemble model. The ensemble model stacks the predictions from various individual models, and gives a weighted average of all predictions. In my analysis, the ensemble model has performed the best among all the models as the prediction errors are the lowest. The econometric models have higher error rates than the machine learning models. In my analysis, I have elucidated each model mathematically and made plots to compare forecasts from each model. Lastly, I have stated the limitations of their usage and ways to enhance the current model.
The following license files are associated with this item: