Hi all,
I'm trying to explore the use of a random forest to predict stock price through the use of twitter data.
Ideally I would like to filter companies based on market cap and sector and then collect monthly summations of each sentiment metric, which would then be input into the model.
I'm fairly new to coding and have never used Blaze to pull data. Specifically I would like to have a filter that looks something like this below:
twitter_data = odo(psych[psych.sid in companies & psych.asof_date <= end_date & psych.asof_date >= start_date], pd.DataFrame)
- psych is the psychsingnal data base
- companies is a set of equity objects
- start_date / end_date are datetime objects
For the life of me I cannot find how to do so on any online source - thank you for any input!