See updates and corrections below…
My apologies to Johnny Mathis. I’ve got Christmas on the brain.
It has been on the to do list for some time to begin investigating the affect of volume. With the S&P 500 extended and making 20 day highs, I decided to start with a simple investigation.
Rules:
- Buy SPY at the close when it makes a 20 day high OR
- Buy SPY at the close when it makes a 20 day high with volume LESS than the 50 day average OR
- Buy SPY at the close when it makes a 20 day high with volume MORE than the 50 day average.
- Sell X days later at the close (up to 20 days later).
Results:
Vol<50DMA under-performs buying all 20 day highs (regardless of volume), but not by much.
Perhaps the most interesting result is the performance of buying the 20 day high with volume>50DMA.
As volume is running far below the 50 day average (as is normal for the season) and SPY is making new 20 day highs, I expect that we might see some consolidation or weakness once the positive seasonality wears off.
UPDATE:
See Toptick’s observation in the comments section.
I ran the tests again and pasted the results into the same spreadsheet. I’m not sure what I did wrong previously (I suspect I moved the wrong column into the graph), but the current graph is now correct.
Hi, Wood:
Maybe I’ve already gone on holiday, but something bothers me about the chart. By definition it would seem that the set of returns for “buy 20 day high” is the union of the sets of returns for “buy 20 day H vol > 50 day” and “buy 20 day H vol < 50 day", so the average of returns of "buy 20 day highs" should be a weighted average of the subset averages, i.e., the blue line should always be between the red and green lines. At 20 days out, V50DMA days looks like 0.32%, so how do all those returns average 0.51%? What have I misapprehended?
Thx!
Top, I noticed that too. I set the post to post automatically (I wrote it several days ago) and don’t have instant access to AmiBroker. I’ll take a look in Amibroker tonight and see if I can figure out what is going on.
Top, good call. The graph is now correct. Thanks!
Mr. ‘Shed,
Have you tried this with a 20 day low? I am curious to see your conclusions on volume when you’re done with these posts. I haven’t tested volume before, but my gut says it means less than we are thought to believe (at least for shorter term mean reversion trading). But my gut is wrong a lot ;). I always appreciate your insightful posts. Enjoy your holidays!
-Chris
Chris, I am anxious to try the 20 day low. Will try that soon! Happy Holidays!
Hi Wood,
Another PPT member recommended that I try your trading system due to my very limited time for trading in the coming year. While looking up some information in the “learn more section” I am not sure I got the fact right. One chart shows that all trades average gain 1%+. How does that compound to the huge gain from $100K to $2.6millon? Does that mean all $100K is invested at all time and gain 1% each week for the 50 weeks or so? Is there a possibility that I pick only the 30% losing trades? And if I am busy during the week and can’t respond to mid-day trading adjustment, would all preset buy/sell orders do to make reasonable gains?
Thank you.
Hubbs, I sent you an email. Let me know if you don’t get it.
Merry Christmas Wood and thank you once again for this great trading system.