Because of my own holiday schedule, I did not update the tables the past weeks. I will be backfilling the data going into next year while also providing more formats and better visualization tools.
This project publishes a weekly correlation matrix of all US equities with daily trading activity. The correlation analysis is based upon the equity returns of the last 50, 100, 200, and YTD trading days. Underlying price data is sourced from a premium market data vendor and correlations are calculated on split/dividend adjusted log returns.
Every Saturday, four sets of correlation matrices are created, archived, and uploaded containing the 50, 100, 200, and YTD equity correlation matrices. The git repository only functions as a download URL mirror and index to the externally hosted files. Each 7zip correlation matrix archive contains three files, the actual correlation matrix, a list of symbols contained in the respective analysis, and the date range of the calculations.
The compressed files and their contents use the following naming pattern:
enddate_window.ext
20151106_200d.7z
20151106_200d.csv
20151106_200d_symbols.csv
20151106_200d_dates.csv
20151106_YTD.7z
20151106_YTD.csv
20151106_YTD_symbols.csv
20151106_YTD_dates.csv
The data CSV files contain large square correlation matrices. The header row is a comma delimited list of symbols, where the first header field is empty. The first field in each row is an equity symbol followed by the corresponding correlations of that symbol to the column symbols. Example:
| AAPL | GOOG | IBM | QQQ | SPY | |
|---|---|---|---|---|---|
| AAPL | 1.00 | 0.41 | 0.53 | 0.76 | 0.70 |
| GOOG | 0.41 | 1.00 | 0.45 | 0.67 | 0.56 |
| IBM | 0.53 | 0.45 | 1.00 | 0.67 | 0.72 |
| QQQ | 0.76 | 0.67 | 0.67 | 1.00 | 0.95 |
| SPY | 0.70 | 0.56 | 0.72 | 0.95 | 1.00 |
> df = pd.read_csv('20151106_YTD.csv', index_col=0, header=0)
> df.AAPL.GOOG
0.40999999999
If you have comments or issues, please [file an issue] (http://github.com/astocko/equities_corr/issues/new) or contact me on twitter, @astocko.
US Equities Correlation Analysis by Alexander Stocko is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.