Deep Learning Setup For Dow-30 Stocks

DLPAL workspace and deep learning results for DOW-30 stocks as of the close of Friday, October 21, 2016.

For  this setup we need adjusted data for all Dow-30 stocks since 01/2000. The DLPAL p-indicator workspace setup is shown below. More information about the p-indicator can be found in the program manual.

p-dow_wks_20161021

We have applied a profit target and stop-loss of 2% because we are interested in short-term directional price action. We also marked “Show All results” because we then would like to calculate the P-Dow indicator value. The longer-term trend is removed from the results by checking “Detrend All results.

Results

p-dow_20161021

After we click Stats we get a P-Dow indicator value of 0.70 (number of stocks with positive directional bias divided by all stocks in results.) This is a relatively large directional bias and indicates a higher probability for an up day. Please note that this is probability, not a prediction.

In addition, WMT and PG have high directional bias P-Long and P-Delta (P-long – P-short) at high significance S above 10. DIS has high negative short-term bias at high significance.

Note that use of a larger universe, such as S&P 500 stocks, can facilitate execution of long/short equity strategies by selecting stocks at the top with high probability P-long and stocks form the bottom with high probability P-short, both groups at a high significance levels.

If you have any questions or comments, happy to connect on Twitter: @priceactionlab

You can download a demo of DLPAL from this link.

Subscribe via RSS or Email, or follow us on Twitter.

Disclaimer

Copyright Notice

This entry was posted in Machine learning and tagged , , , . Bookmark the permalink.

Leave a Reply