Search found 68 matches

by Eventhorizon
Tue Mar 06, 2012 7:16 pm
Forum: Testing and Simulation
Topic: annual profit target
Replies: 16
Views: 5864

Babelproofreader, you know I would be interested!
by Eventhorizon
Mon Mar 05, 2012 8:16 pm
Forum: Testing and Simulation
Topic: annual profit target
Replies: 16
Views: 5864

Thanks for your kind comments, Ric and Marriot. If I had to guess (and that is all I can do) I would say Winton did the following optimization of their trading system: Minimize volatility subject to achieving a return greater than or equal to the typical pension fund return assumptions. I think Wint...
by Eventhorizon
Mon Mar 05, 2012 10:57 am
Forum: Testing and Simulation
Topic: annual profit target
Replies: 16
Views: 5864

Here's one I prepared earlier ... Here's an analysis of the change in performance of Winton's Diversified Fund over the last 14 years which I did as an illustration of a Change Detection Algorithm. Look near the beginning of this post where I set up the problem . Look at the last section in this pos...
by Eventhorizon
Thu Mar 01, 2012 11:36 am
Forum: Testing and Simulation
Topic: Should one calculate ATR based on the back-adjusted series?
Replies: 2
Views: 1359

Depends how you did your back-adjustment. If you used a Panama approach (i.e. adjusting each contract up or down by addition) ATR is same for the actual and back adjusted series: Adjusted Price = Actual Price + delta => adjusted ATR = Actual ATR If you used a ratio approach, to get the traditional A...
by Eventhorizon
Thu Feb 16, 2012 11:49 am
Forum: Testing and Simulation
Topic: Sharpe Versus Annual Sharpe
Replies: 7
Views: 2204

Upon waking this morning, I realized that the most likely source of differences in the Sharpes is going to be in the risk-free rate if the averages and standard deviations agree - doh! These years of zero short-term interest rates are having some effect - on my brain! Jake, I would definitely apprec...
by Eventhorizon
Thu Feb 16, 2012 1:07 am
Forum: Testing and Simulation
Topic: Sharpe Versus Annual Sharpe
Replies: 7
Views: 2204

trender41, I am not sure how deep into this I want to get, but there are issues with some of the statistics produced by TB. I have a long-standing project under development: a TB statistics audit block. The idea being every time you upgrade to a new version, you throw this block into your system and...
by Eventhorizon
Tue Feb 14, 2012 5:04 pm
Forum: Trend Indicators and Signals
Topic: Bloomberg Trender Indicator
Replies: 7
Views: 7002

The analysis might be thrown off by the initialization of the EMA's. I don't know how many bars of data you are fitting to - this effect becomes less significant the more bars you have. Possible ways of dealing with this: 1) Perhaps if you introduced two additional variables, the initial values of E...
by Eventhorizon
Sun Feb 12, 2012 1:34 pm
Forum: Trend Indicators and Signals
Topic: Bloomberg Trender Indicator
Replies: 7
Views: 7002

I was curious so I looked at the pdf (linked by Babelproofreader) describing the indicator. Since the formula for the indicator is specified, I assume that it is the parameters you are trying to figure out. Here's some things I noticed: 1) The diagram on page 2 must be a conceptual representation ra...
by Eventhorizon
Tue Jan 31, 2012 12:02 pm
Forum: Testing and Simulation
Topic: Musings on (worthless) Past Performance
Replies: 83
Views: 21293

Dean, There is something funky with those numbers. For example, if the data is for 5 years of returns, there are several entries where the compounded return is positive but the Sharpe is negative. Since arithmetic return is always greater than geometric (i.e. if compounded return is positive then ar...
by Eventhorizon
Sun Jan 29, 2012 1:43 pm
Forum: Futures Markets
Topic: Eurodollar (ED) - limited upside - PART 2
Replies: 8
Views: 2521

My understanding of ED is the settlement price is 100 - 3 month Libor. i.e. as rhc says, if 3 month Libor = 0% at expiration, then the settlement is 100. So, unless Libor goes negative, ED is capped at 100. Question is, can ED Libor go negative? I searched around and could only find examples of Swis...
by Eventhorizon
Fri Jan 27, 2012 4:04 pm
Forum: Testing and Simulation
Topic: Traders, learning to write computer programs
Replies: 20
Views: 7736

In addition to Matlab / Octave, I would recommend R, in the role suggested by PNW_Trader. It's open source, there is a huge library of gnu packages that comes with it, a lot of focus on financial time series analysis, etc. It seems, to me at least, to be the dominant statistical package in academia....
by Eventhorizon
Tue Jan 24, 2012 3:33 pm
Forum: Testing and Simulation
Topic: Following your system . . . . with a delay
Replies: 18
Views: 5973

Now if you are in the original trade and it moves against you then how is this different to entering the trade a bit later on at the risk you can afford. It's a variation on the Monty Hall problem - the trade moving against you is like one of the doors being opened: you have posterior information. ...
by Eventhorizon
Tue Jan 24, 2012 1:22 pm
Forum: Testing and Simulation
Topic: Following your system . . . . with a delay
Replies: 18
Views: 5973

I would say it depends on what information you have about your system and how it trades in terms of probabilities. Beware of the possible implications of what you find. For example, if the expected return of a trade entered with a delay is higher than that of a trade entered immediately, should you ...
by Eventhorizon
Mon Jan 09, 2012 12:19 pm
Forum: Money Management
Topic: Is Modern Portfolio Theory Dead?
Replies: 22
Views: 14646

Nice work stopsareforwimps. Here's something similar I did a while ago. Almost the same idea (contracts filtered and ordered by volatility on the vertical axis, dependent contracts on the horizontal). I used a different set of contracts and looked at only 3 cases: 5% most volatile days upwards, down...
by Eventhorizon
Tue Dec 13, 2011 1:34 pm
Forum: Brokers
Topic: Interactive Brokers & Re-hypothecation
Replies: 19
Views: 19284

I find ZH's blog site to be fun and all, but whoever writes it has a penchant for hyperventilating. Let's be clear, without hypothecation you couldn't extend margin loans to clients. Do you expect the broker to lend their own capital to you so you can buy assets with their money? Of course not, they...
by Eventhorizon
Wed Nov 02, 2011 3:57 pm
Forum: Testing and Simulation
Topic: Contract month selection and rolling parameters
Replies: 14
Views: 5381

Bravochico, Would it be possible for you to cobble together an example that shows how using raw contracts would give a different back-test result from using continuous contracts? I am really interested to understand the mechanism, and whether the issue is just a case of using the "wrong" type of con...
by Eventhorizon
Wed Oct 26, 2011 12:42 pm
Forum: Trend Indicators and Signals
Topic: Why spending too much time on forecast?
Replies: 5
Views: 3650

moto moto,

Your corns are prescient - we have 4" of snow and it's still coming down!
by Eventhorizon
Thu Oct 20, 2011 12:01 pm
Forum: Testing and Simulation
Topic: Turtle System for Tradestation 9 Easy Language
Replies: 16
Views: 7867

I think there are two approaches to this ... 1) If you run a back-test today and get a certain set of trades, then run a back-test at a later date after the current set of front-month contracts have expired then the back-adjusted contract values will have changed due to the rolls taking place betwee...
by Eventhorizon
Thu Oct 20, 2011 11:49 am
Forum: Trader Psychology
Topic: Current Market Psychology
Replies: 11
Views: 5584

Hi Rich,

Nice to hear from you ...

I have always felt that your great-grandfather + theory explains the persistence of the Kondratieff Cycle.

to echo rhc: history doesn't repeat but it rhymes (Twain)
.. though I have a hard time coming up with anything that rhymes with "history"!
by Eventhorizon
Mon Oct 17, 2011 8:05 pm
Forum: Testing and Simulation
Topic: Anyone using MatLab, Octave, SciPy, the Language R
Replies: 22
Views: 12003

I just use R for off-line analysis - it is not suited to building a trading system as it is horribly slow for loop processing i.e. working through this time interval then the next and so on - you cannot vectorize in this case because of the time-series dependencies so why use R?