Search found 68 matches

by Eventhorizon
Sat Mar 26, 2011 1:21 pm
Forum: Testing and Simulation
Topic: Another new Goodness Measure; also, walking it forward
Replies: 18
Views: 6636

Sluggo,

Nice version of the chart - much clearer than mine!!
by Eventhorizon
Thu Mar 24, 2011 5:46 pm
Forum: Testing and Simulation
Topic: Another new Goodness Measure; also, walking it forward
Replies: 18
Views: 6636

Thank you Sluggo for sending me off on an interesting detour!! I realized what I had written above regarding the pareto frontier was not accurate, so I am hoping to rectify that with this post. Following is an idealized pair of objective functions (just parabolas). One might be, say, MAR the other C...
by Eventhorizon
Mon Mar 21, 2011 5:10 pm
Forum: Testing and Simulation
Topic: Another new Goodness Measure; also, walking it forward
Replies: 18
Views: 6636

Thanks for the interesting links, Sluggo, and the history of Lake Ratio. Our present problem is, generally, to find that single set of parameters that maximizes our bliss. For the benefit of those not familiar with the field (including me), the solution to multi-objective optimizations is a solution...
by Eventhorizon
Mon Mar 21, 2011 11:22 am
Forum: Testing and Simulation
Topic: Another new Goodness Measure; also, walking it forward
Replies: 18
Views: 6636

Rabidric, I have been experimenting with integrating the draw-down over the duration of the back-test to arrive at E(DD) i.e. the mean state of draw-down if one picked a day at random from the back-test. If you combine this with regressed annual return (divide RAR% by E(DD)) one arrives at a nice pa...
by Eventhorizon
Sun Mar 20, 2011 12:37 pm
Forum: Testing and Simulation
Topic: Another new Goodness Measure; also, walking it forward
Replies: 18
Views: 6636

Bobsyd, You have raised a quite fascinating issue. First I want to make sure I have correctly understood it! Concept: Maximizing one goodness measure, Gw, in a walk-forward simulation may lead to the global maximization of another goodness measure, Gg, across the entire span of the simulation. In th...
by Eventhorizon
Fri Mar 11, 2011 8:46 pm
Forum: Testing and Simulation
Topic: Another new Goodness Measure; also, walking it forward
Replies: 18
Views: 6636

Let's use %RAR and we could call it the Robust Rabid Ric Ratio (R^4)!
by Eventhorizon
Wed Mar 09, 2011 10:06 am
Forum: Money Management
Topic: How to incorporate subscriptions in trading
Replies: 14
Views: 6201

Ace, surely you can back-test this. Why not explore what would have happened over your back test period to a hypothetical capital addition - all you need is the daily value of your total equity curve, and some estimate of slippage in adding to open positions. Then you can provide your investors with...
by Eventhorizon
Mon Mar 07, 2011 2:23 pm
Forum: Money Management
Topic: How to incorporate subscriptions in trading
Replies: 14
Views: 6201

Ace, I think there is really only one decision here: Is it better to scale up already open positions or should you wait for new trades to come along which you size according to the new amount of AUM. You had to make this decision the day you started trading - you went from zero AUM to $x AUM; did yo...
by Eventhorizon
Mon Feb 28, 2011 11:20 am
Forum: Testing and Simulation
Topic: Profit Target exits improve this system's performance
Replies: 63
Views: 30583

stopsareforwimps ...

Google + "Hakansson" gets me stuff about cosmetics and some steel mill dude!

Google + Hakansson + Bermuda Triangle gets me your post! LOL.

I am curious to follow up, can you be more specific about Hakansson, or am I being dense (it happens often)?
by Eventhorizon
Mon Feb 28, 2011 11:13 am
Forum: Testing and Simulation
Topic: Blending noncorrelated (or anti-correlated) equity curves
Replies: 50
Views: 30047

Mark, there are some things going on here that are worth drawing attention to: the time frames of one's analysis are a key element here. An error I continue to make because this stuff is not intuitive to me yet, is to assume that because two instruments are correlated, they move the same direction. ...
by Eventhorizon
Mon Feb 28, 2011 10:50 am
Forum: Testing and Simulation
Topic: EasyLanguage
Replies: 1
Views: 1332

Seems to ... you can verify by putting the following into Script: Entry Orders: IF instrument.symbol = "YourSymbol" THEN PRINT "tradeDayOpen", instrument.tradeDayOpen, "today's open", instrument.open ENDIF Put a de-bugging break on the print statement. You should see tomorrow's open being returned a...
by Eventhorizon
Sat Feb 26, 2011 1:43 pm
Forum: Data Providers and other non testing software
Topic: Futures data source (individual contracts, term structure)
Replies: 1
Views: 1541

Thank you, Eric.

That's a handy resource.
by Eventhorizon
Mon Feb 21, 2011 7:05 pm
Forum: Testing and Simulation
Topic: Blending noncorrelated (or anti-correlated) equity curves
Replies: 50
Views: 30047

By way of explanation ... the differences in solution arise because I simulated 2 series of length 1000 with the characteristics mentioned above and jammed them through some code I use to calculate the eigen-decomposition of a covariance matrix of real multiple data series! A sledgehammer compared t...
by Eventhorizon
Mon Feb 21, 2011 5:25 pm
Forum: Testing and Simulation
Topic: Correlation Matrix:
Replies: 11
Views: 5092

It is trivial to use R to open your actual futures files and generate a complete correlation matrix and display it as a heat map. With a few lines of code you can do a Monte Carlo analysis to estimate VAR. It is worth examining the covariance of the outliers to get a sense of what happens when the m...
by Eventhorizon
Mon Feb 21, 2011 1:09 pm
Forum: Testing and Simulation
Topic: Blending noncorrelated (or anti-correlated) equity curves
Replies: 50
Views: 30047

LeviF wrote: There must be some other factors at play here aside from just equity curve correlation. I can come up an uncorrelated system with negative expectancy that hurts overall portfolio performance. Or what about two systems with -1 correlation (system 2 takes the opposite trade of system 1). ...
by Eventhorizon
Sun Feb 06, 2011 11:11 am
Forum: Data Providers and other non testing software
Topic: csi advanced backadjuster
Replies: 12
Views: 4892

Chris, I am not familiar with the full-blown CSI UA package, I am currently testing using the "free" data made available here. I know you have posted on this before but I can't remember the details of the entire saga, so ignore me if I am asking the obvious ... Where are you seeing the discrepency m...
by Eventhorizon
Mon Jan 31, 2011 1:27 pm
Forum: Testing and Simulation
Topic: Showing your entry parameters
Replies: 1
Views: 1153

I think using the term "parameters" in the TBB context can be confusing - the term is the same as that used for the variables that are set by a user for a given test. The following assumes you mean "indicators". The best way to show "intermittant" indicators (i.e. those whose values are associated w...
by Eventhorizon
Wed Jan 19, 2011 3:22 pm
Forum: Testing and Simulation
Topic: Harmony Search Algorithm
Replies: 10
Views: 8771

Well, here it is. The zip file contains a list of files, and some instructions in pdf form. I apologize in advance for the amateurish nature of my coding and documentation - I have not yet learned the discipline needed to write beautiful code like Roger Rines! The system that is included (thanks for...
by Eventhorizon
Tue Jan 18, 2011 12:10 pm
Forum: Testing and Simulation
Topic: Harmony Search Algorithm
Replies: 10
Views: 8771

Harmony Search Algorithm

In the Feature Requests forum, Bobsyd recntly requested some intelligence on the part of TBB in optimizing system parameters . This got me researching search algorithms. I have successfully implemented a genetic search algorithm called the Hamony Search Algorithm in TBB. I would like to share it wit...
by Eventhorizon
Tue Jan 18, 2011 11:41 am
Forum: Testing and Simulation
Topic: Adjusting the exit depending on time
Replies: 1
Views: 1113

Use instrument.unitBarsSinceEntry[n] to identify how long you have held a unit "n" if you want to count by bars, or use instrument.unitEntryDate[n] if you want to count by calendar days.