Sunday, October 21, 2012

System Update - Profit Curves

Al asked for some up-to-date profit curves.  Ask and you shall receive.  I ran them back to the beginning of 2012.  For curves going back further, please check the "Updates" link above.  I used TNA to calculate the curves.  Using XIV would've resulted in slightly better returns, but I'm too busy to run those numbers.  As I mentioned before, I'll start tracking volatility trades in the official stats next year.

To summarize, the JTS (which is my baby, so to speak) is performing the best.  I really see only 2 poor months - April and May - when the market experienced some turbulence. Since the end of May, the system has performed better than I could've hoped. One could argue August was sub-par, but August was extremely boring.

The CCI has had some issues this year, but is still performing admirably.  Sustained moves in 1 direction with no mean-reversion will hit this system hard.  It thrives in a mean-reversion environment, and did well February through July.

The JT Composite system, which is simply a combination of the 2 above, is what I follow most closely.  It reduces risk while offering handsome rewards.  With such a nice, relatively smooth rising profit curve, I see no reason to deviate or doubt the system.

Ok, enjoy the evening and I'll see you tomorrow! 
-J

(click to enlarge)




12 comments:

  1. Yup, smooth-n-steady is where it's at. Nice work. Would be great to see at least the composite system back to 1998. That'd put you through 3 bull and 2 bear market cycles, including relentless trend, mean-reverting, and parabolic moves. 2012 was pretty boring, overall, frankly, and, for me at least, probably the most difficult year I've seen for trading going back to the early 1990's, so, it was kinda unique, in that regard. Would hate to think you've optimized a system for a statistically unusual year -- I doubt, but would be worth seeing the curves "through thick and thin".

    ReplyDelete
    Replies
    1. Well, one could argue even if I had smooth and steady profit curves back to 1998, there's no guarantee they would continue that way in the future. I'm basically flying by the seat of my pants here only having data back to early 08. But I have to say, so far - so good. I'm going to keep the faith as long as curves continue to look like this year after year.

      Delete
  2. Thanks J. As always, I appreciate how responsive you are to follower's questions.

    Now, to be honest, I was actually interested in the profit curves before 2012. Specifically, I'm curious about how the JTS looks during the period when the CCI system was pretty flat. I'm concerned about the CCI's performance from 2008 to 2011. It seems to have been flat during that period and only come to life this year. Do you think that it's possible that the CCI system and/or the Composite has been particularly attuned to the current market but not to other kinds of market activity? As someone who walks the walk, i.e., who invests his own money in the system, I would imagine that you've thought about these issues, so I'm very interested in your views.

    ReplyDelete
    Replies
    1. Well, no worries...I'm glad I put together these 2012 curves. Should make year-end stats a piece of cake.

      I will take a look at all profit curves here in a bit, but you bet I think about the things you mention. How worried am I that I can't produce a JTS profit curve prior to 2009? A little, but only a little. I added the CCI system as a complementary system to the JTS mainly in order to increase confidence and reduce risk. I figure if a system starts to misbehave, I have a 2nd one to fall back on, so to speak, via the JT Composite. Could both systems fail...sure, but I suppose it's more likely only 1 would fail, and it's a risk I'm willing to take. And as you said, I put my money where my mouth is, and I wouldn't be doing any of this if I didn't believe in it.

      Based on how the the CCI is set-up, I can assure you it will struggle in a strong trending market with minimal daily swings. I can tell you that right now without looking back at 20 years of data. It's sole purpose is to capture the chop offered by equity markets. I'm counting on there being enough choppiness in the future that the good trades outweigh the bad. If I'm wrong, the JTS/JTComp should help minimize losses.

      All my systems have proven to perform better than my past discretionary trading. So I'm better off in that respect. :)

      Delete
  3. To be clear, my basic question is whether the JT System has performed better over time than the JT Comp. In other words, do non-mean reversion periods drag the Comp down sufficiently (because of the relative weakness of the CCI system during such periods) that its performance is not up to that of the original JT system?

    ReplyDelete
    Replies
    1. Yes, most times the JTS will perform better than the JT Comp. The reason is the JTS is a far more superior/robust model than the CCI.

      So some may wonder why I don't trade it exclusively and the answer is that I think the CCI offers some value. It also adds confidence to my trading. I enjoy seeing them in agreement and will feel more comfortable placing my money on the line. And when they are opposite, I get cautious and force myself to the sidelines, a la the JT Comp.

      Now, could the JT Comp be tweaked to weight the JTS more than the CCI. Sure! Using the current example (JTS-short,CCI-long) I could be holding a small short position right now. I guess out of simplicity, I gave them equal weighting in the JTComp. Perhaps this winter I'll take a look at a 2:1 weighting ratio and run the Composite numbers again. Thanks for the insightful questions.

      Delete
  4. Wish we would have stayed short... markets are plumb out of strength...

    ReplyDelete
  5. You're right, RMI, and the original JT system is short.

    But watch the VIX. It's currently above its Bollinger Bands. If it closes above them, that's often a sign of an impending bounce in equities.

    ReplyDelete
  6. J, thanks for the thoughtful replies.

    Here's one more request. I tried to find this data on the blog, but I failed. Could you briefly post what the monthly or annual returns for the three system would have been since their inception?

    Thanks.

    ReplyDelete
    Replies
    1. I'll dig the info up, and I'll make a note to include it in the year end stats post as well.

      Delete