Dynamic Portfolio Balancing Project

Discussions about Money Management and Risk Control.
Post Reply
jklatt
Roundtable Fellow
Roundtable Fellow
Posts: 79
Joined: Sat Feb 17, 2007 11:51 am

Dynamic Portfolio Balancing Project

Post by jklatt » Tue Sep 16, 2008 5:58 pm

I thought I'd post a project I'm undertaking on this board. It's a work in progress and mainly just an exploratory exercise to see what can be done mixing and matching equity curves and adjusting position sizes dynamically. I'm posting some of this here as I'm hoping to get some constructive feedback that might stimulate new ideas. Who knows where this might go, right? The sky is the limit.

I've spent a few years, off and on, designing and testing systems in Matlab. I've yet to come up with anything that really knocks my socks off. I can get MAR ratios over 2 for 20 year periods, but I feel like I get some of those ratios because of the frequency of the system I choose and the portfolio I choose -- not because there is anything special about what I've designed. Most of these systems are moving average based, time based breaks outs, volatility based breakouts, etc. -- trend following to a large degree.

I've noticed that typical trend following systems with typical portfolios over the last few years have become increasingly volatile. From what I've noticed in my tests, it is mainly because a lot of these instruments have become more noisy (more noisy at the entry points and less follow through) and they've correlated with each other to a much larger degree in the past. The most common approach I've seen to combat this problem is to add different markets to the portfolio. I don't like this approach as I think it is just a band-aid instead of a "real" fix for the problem. I'd rather try to take a "lousy" portfolio and try to make the performance better by adjusting my portfolio and position sizing, but I guess that's just my opinion.

So what I want to do is the following:

1) Take a relatively "lousy" trend following portfolio (something that did pretty good in the 90s but got hit with degrading performance in terms of return and drawdown in the recent years).

2) Generate a lot of different equity curves by applying trending following techniques to each instrument of varying degrees of frequency (think 10 day breaks, then 20 day breakouts, then 30 day breakouts, etc.).

3) Take all of these equity curves and try to dynamically mix and match them at different times and weightings to produce a "superior" (in terms of return and drawdown) equity curve.

This is a "proof of concept" project and some of what I might do in my testing might be difficult to implement in real time (resizing on the close using information from the current day -- things like that). If I find that the results have some promise, I would then look at ways to adjust what I'm doing to make it easier to implement.

Again, this is a work in progress. I'll report back what I do and what I find and please feel free to give any feedback you might have as that is what I'm hoping to get by posting this project here.

Jason

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Tue Sep 16, 2008 7:12 pm

These markets were once high flyers but have underperformed recently, which means they fit your requirements quite well:
  • Live Cattle
  • Orange Juice
  • Wheat (CBOT)
  • Silver
  • Cocoa
A different approach is to trade a huge portfolio but limit the total amount of risk exposure at any one time. If you get a new entry signal when you're already maxxed out on risk exposure, you could skip new entries (a la Turtle), OR you could toss out an old trade to make room for a new one (a la the Parking Lot), OR you could resize all existing positions to be smaller, creating headroom, OR (insert your idea here).

You could measure risk exposure by counting the number of positions (a la Turtle), OR by adding up the total-risk-as-a-percent-of-total-equity (a la Heat Limiter systems), OR (insert your idea here).

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Re: Dynamic Portfolio Balancing Project

Post by LeviF » Tue Sep 16, 2008 7:43 pm

klatt_attack wrote:The most common approach I've seen to combat this problem is to add different markets to the portfolio. I don't like this approach as I think it is just a band-aid instead of a "real" fix for the problem.
Its tough for TFs to scrape out a living in volatile, trendless markets, no matter the timeframe. Adding additional markets may in fact be the "real" fix you are truly searching for.

jklatt
Roundtable Fellow
Roundtable Fellow
Posts: 79
Joined: Sat Feb 17, 2007 11:51 am

Post by jklatt » Tue Sep 30, 2008 5:00 pm

It has been a couple weeks and I haven't really come up with anything interesting to discuss. The part I'm getting stuck on is finding different entry and exit techniques that produce "uncorrelated" outcomes. Simple cycling through the parameters of a 10, 20, 30, 40... day breakout system really doesn't do the trick. So I've been stuck on looking for "out of the box" entry and exit strategies which has kind of gotten me off course a bit. Maybe I should take wanting to find uncorrelated trade outcomes to my tribal council.

As for limiting overall/sector heat, I've tested that on a run of the mill portfolio and what I've found is that rejecting trades on this basis has helped quite a bit over the past few years at the cost of gains in the 80s and 90s. I think I've figured out why you could run relatively high sector/overall risk in the 80s and 90s and do rather well and limiting portfolio/sector heat by itself doesn't seem to be the complete solution. If you think of your open positions as a soup, it matters less upon how much soup you're making rather than how the soup tastes (if I add this new position, how will that change how the soup tastes and will it still taste desirable?) Maybe that doesn't make sense. Probably not. I haven't been known to have a way with words.

The reason why I'm a bit skeptical of trading huge baskets of markets isn't because I don't think it's a good idea (I think it is), but doing so without mastering how to trade the run of the mill basket will still leave you open to the problems that have arisen over the last few years (markets have become more noisy and much more correlated). What happens when everyone figures out that this or that foreign market has been smooth and profitable over the past decade? Everyone will pour into the market and turn it into a much less desirable area to trade and if your system can't handle that (like a lot of systems over the past 5 years or so with run of the mill portfolios), you're leaving yourself open to potentially rough times yet again.

jklatt
Roundtable Fellow
Roundtable Fellow
Posts: 79
Joined: Sat Feb 17, 2007 11:51 am

Post by jklatt » Sun Jan 04, 2009 10:40 pm

I've been working on things off and on as usual. I've had a lot of problems finding edges in higher frequency breakout systems and my original intent was to have a mix of high and low frequency systems and mix and match the equity curves. So on that front I've been unsuccessful.

This weekend, I sat back down at the computer and did a bit of "data mining".

I took a "lousy" portfolio (which consists of a lot of meats, softs, metals with a few currencies and interest rates thrown in) and applied my "price strength" indicator to the various instruments.

I set a minimum and maximum indicator frequency and stepped through the range. I then divided the indicator readings into bins (think 0 to 10% stochastic reading, 11 to 20% stochastic reading, etc.) and populated each bin with the next day's return expressed in today's 10 day ATR. If someone could give me some direction as to how to post a picture here, I'll put a pic up of the results.

What I found was that for every frequency selected, the vast majority of the gains (35%+) came when the strength indicator was in the top 2 "strongest price action" bins and that price spent about 17% of the time in these 2 bins.

Something else that was interesting was that there was positive expectancy (going long was profitable) when price was in the top several "weakest price action" bins. This mean that buying weakness for this portfolio and constantly readjusting position size the current 10 day ATR was profitable. I'm not sure why this is the case. Maybe it is a programming error but I don't think it is. I think it's more of a product of the portfolio itself gaining roughly 800 ATRs in 25 years (bull market in general) and the fact that as prices grind lower, ATR tends shrink and when price finally bottoms, prices tend to "explode" higher. This really makes me ask the question, what's the value in trying to go short unless it's in an effort to smooth out your equity curve?

Anyway... I thought I'd just check in and throw up another post on what is going on with me regarding this ongoing project. From here, I plan to put together some equity curves using the following logic:

1) Only enter long when the highest frequency strength indicator is in one of the top 2 strongest bins. Exit the position if strength falls X amount of bins. This will help to keep my "stops" tight.

2) As price moves upward and the lower frequency strength indicators reach "top 2 status", shift my stop to X amount of bins decrease at the new frequency.

I'm not sure if that reads clear but basically the idea is to only go long, start with a "tight stop" and as price moves in favor, expand the stop distance.

I'll post the results if/when I get around to doing this.

jasonz
Full Member
Full Member
Posts: 10
Joined: Sat Apr 14, 2007 10:37 am

Post by jasonz » Thu Jan 08, 2009 3:36 pm

Just posting to say thanks for your posts and I'm interested in following what you are doing.

jklatt
Roundtable Fellow
Roundtable Fellow
Posts: 79
Joined: Sat Feb 17, 2007 11:51 am

Post by jklatt » Fri Jan 09, 2009 12:53 am

Thanks for the kind words... I wish I was more energetic about the project. Maybe I'm really just making life difficult by limiting myself to "lousy" markets but for whatever reason I feel like if I can get past the "profitable/flat in 90% of the lousy markets" hurdle, the end product will be something special when I add the historically big trenders into the mix and start working on risk control, etc. I envision a winning % greater than 50% and a win/lose size ratio of 2+. Right now, that feels like a pipe dream!

Post Reply