I have basic knowledge in time series prediction and supervised/unsupervised machine learning algorithms (clustering, classification, decision tree, etc.) I am now given a task to predict a bunch of stock prices. Each stock has its previous trading price (a period of 18 months) as well as some other features: coupon, asset rating, industry, etc. I only know how to use time series analysis or supervised machine learning separately, I have no idea of how to combine these two together. Is there any particular algorithm that I can use as a predictive model? What are steps to combine both dynamic and static information? Any help will be appreciated!
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
It depends on what you mean by "combine these two together". Predicting the stock price with a suffiently large dataset sounds like pretty standard application for common time series models like ARIMA (I know in finance GARCH is pretty common as well but I don't know if this is applicable here). So this would be similar to building a regression model for prediction purposes. However also algorithms from supervised learning can be applied to time-series data like random forests or neural networks.
