Testing correlation between stock price change and financial reports

A small script to test if there's a correlation betweeen the change in the stock price on the day when the company releases its financial reports:

• Balance Sheet
• Income Statement
• Cash Flow Statement

To give a concrete example, this tries to test the probablility that a stock price will increase on the day when the company releases its financial reports and one of the values in these reports shows an unexpected increase / decrease since the last year.

Note that the stock price change is calculated as, stock_price_on_day_of_release - stock_price_on_next_day_after_relase.

Disclaimer

Open a PR if you see a mistake from my part.

Usage

Financial modeling prep is used to gather the data, thus you need to have an API key from them. You can get one here. The free version won't be able to gather all of the tickers in one run, because roughly 7 requests are needed for 1 ticker.

num_tokens is used to run through the script quickly, if needed.

# run at first to gather the data
python3 main.py --token YOUR_TOKEN --num_tickers 500 --mode FETCH
# run to calulate the correlation on the saved data.
python3 main.py --token YOUR_TOKEN --num_tickers 500 --mode CORR

Results

All of the correlation results can be found in the data/correlation.csv file. 2389 years of reference were used to find these results.

The highest correlation can be found between the stock price change and otherLiabilities_pct_change (-0.15). Other than that, the correlation is not significant.