The project aims at developing a model to predict the VIX. The dataset includes a variety of data with different categories.
Things I like:
It collects a very rich dataset that includes plenty of financial and economic data. All data are meaningful and relevant to make predictions on VIX.
It makes beautiful visualizations on data descriptions. The correlation analysis helps me better understand the relationship between different features.
It has used decision tree model to make a preliminary analysis and it also ranks the features according to their importance.
Things needed to be improved:
The pictures in the report are too small to be clear. It would be better to include all the pictures in your Github folder.
It seems arbitrary to group the VIX data into three classes in the way the project specifies. I suggest to give reasons and supports why it should be grouped in that way.
There are no explanation why selecting the features with feature importance larger than 0.08. Maybe it needs cross validations to select a more reasonable critical value.
The project aims at developing a model to predict the VIX. The dataset includes a variety of data with different categories.
Things I like:
Things needed to be improved: