David Harding Quote

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
ladidalimey
Roundtable Knight
Roundtable Knight
Posts: 197
Joined: Thu Sep 21, 2006 6:21 am
Location: London

David Harding Quote

Post by ladidalimey » Wed Oct 05, 2011 11:18 am

I thought this was an interesting quote from David Harding
There are no discrete events in our investment process. We do not put on a trade monitor it and take it off. We stopped doing anything like that 10 years ago.
From the bottom of this page..

http://www.thehedgefundjournal.com/maga ... arding.php

I'd be interested to hear the forum views.

Thanks,

L.

svquant
Roundtable Knight
Roundtable Knight
Posts: 126
Joined: Mon Nov 07, 2005 3:39 am
Location: Silicon Valley, CA
Contact:

Post by svquant » Thu Oct 06, 2011 12:10 pm

From an older Winton pitchbook - perhaps this helps? May have an even more detailed presentation from back when they had less than $500MM AUM. Note as they have grown the marketing message and details have changed greatly but I have seen version of this slide in presentations as late as 2009.

While many CTAs do not talk about moving averages as a forecasting methodology, i.e. just look for crossovers, it is a very well developed field. Just look at any time series forecasting text book and you can see how others look at exponential moving averages for example.

So the mechanisms shown by Winton could all be done via single or double moving averages, i.e. well recognized LTTF techniques.
Attachments
Winton Mechanisms.JPG
Winton Mechanisms.JPG (55.21 KiB) Viewed 5835 times

ladidalimey
Roundtable Knight
Roundtable Knight
Posts: 197
Joined: Thu Sep 21, 2006 6:21 am
Location: London

Post by ladidalimey » Thu Oct 06, 2011 12:45 pm

Thanks for that svquant, that's very helpful. If you have any other stuff I'd like to see it.

L.

Chris67
Roundtable Knight
Roundtable Knight
Posts: 1046
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 » Thu Oct 06, 2011 12:57 pm

do remember that David HARDING is a marketer - he is full of it when it comes to what Winton do - trying to convince investors that they are somehow different to others when in fact they are not really different at all as their results clearly demonstrate
Also intyersting chart - I'd like to know ho wthey forecast daily volatility - seems a bit predictive

What they probably have a is a 100/20 Dual MA system and that its - teh rest of what they do is trying to work out how to execute trades
Say they risk 0.25% on a trade and get a sell signal on corn - presumably they have to risk about 100$ Mill on the trade - difficult to see how they can do it ?

If I were an investor the querstion Id be asking Winton is how they intend to deal with all teh new regs on their way limiting speculative position sizes in U.S commodity futures (and presumably Europe to follow) - instead of asking that sort of question I'm pretty sure their institutional client base will instead be far more intrigued to hear about their holy grail techniques that blow teh competetion anway and all their phd's in Oxford
C

svquant
Roundtable Knight
Roundtable Knight
Posts: 126
Joined: Mon Nov 07, 2005 3:39 am
Location: Silicon Valley, CA
Contact:

Post by svquant » Thu Oct 06, 2011 1:44 pm

Chris:

Let me just make a few comments on the technical points you mentioned.

Volatility forecasting is well known and MUCH easier to do in an economically useful manner than return forecasting. Lots of academic and industry papers on this. So forecasting daily volatility in a predictive way is not a big thing in my opinion. Many Blox users maybe doing this without even knowing it. If you use a daily ATR value to adjust position sizes or determine position size on a new trade then you are following the Winton mechanisms to a certain degree.

Perhaps they do correlate highly to a 20/100 DMA simple system - or perhaps they use the value of the 20/100 oscillator (MACD) as a dependent variable in forecasting returns for the next N-days and then do mu-k-sigma portfolio optimization based on the forecasted/estimated covar matrix. Seen it done by some other multi-billion dollar funds at the daily and even intraday basis.

It is very hard to compete with them or BlueTrend if you are a highly correlated trend follower. Perhaps years like 2011 will allow you differentiate as trend followers are all over the place. Just look at Transtrend another trend following juggernaut they are down for 2011 YTD or Chesapeake which is having a very bad year @ -15.9% YTD.

michaelt
Senior Member
Senior Member
Posts: 37
Joined: Wed Feb 03, 2010 10:53 am
Location: princeton,nj

Post by michaelt » Thu Oct 06, 2011 2:38 pm

Hi, may I ask where you get your info for results of traders performance, such as Chesapeake?

rgd
Roundtable Knight
Roundtable Knight
Posts: 396
Joined: Thu Aug 10, 2006 5:21 pm

Post by rgd » Thu Oct 06, 2011 3:22 pm

michaelt - IASG or Autumngold might be a good place to start

I would venture to guess that Winton is doing things that are light years beyond simple DMA systems. I know I would be if I had 100 quants in lab coats in my research dept.

I read a presentation by Winton that had a profound impact on me, the gist of which was:

- strategies with high sharpe ratios are hard to come by and have a limited shelf life
- low sharpe ratio strategies are easy to come by and tend to hold up over time
- therefore, creating and blending a plethora of non-correlated, low sharpe strategies makes for a very nice equity curve

svquant
Roundtable Knight
Roundtable Knight
Posts: 126
Joined: Mon Nov 07, 2005 3:39 am
Location: Silicon Valley, CA
Contact:

Post by svquant » Thu Oct 06, 2011 4:22 pm

...and individual strategies with a high sharpe ratios tend to have very limited capacity while they are good on the shelf or in the market

Don't get me wrong wasn't implying that Winton uses simple MA rules (nor the other funds I was framiliar with) just that simple MA techniques can fit within the mechanism described in the slide.

Actually once again if you believe the pitchbooks have some validity then Winton states they specifically forecast over three different time horizons with multiple models. Additionally, they have also stated they have some fundamental models that are not pure price based although the weight of these is much less than their more traditional models.

Bravochico
Full Member
Full Member
Posts: 24
Joined: Wed Sep 14, 2011 11:46 pm
Location: Southern California, Ca

Post by Bravochico » Thu Oct 06, 2011 4:23 pm

I presented along side David in 2002. He had heat maps showing parameter combinations of moving average systems. I'm sure he has evolved as we all do, but he still a momentum guy. His track record better than most TFs, as he extremely well diversified in the true sense. Like Dalio said, after approx. 15 different return streams, further diversification does little. When I hear guys running 50 systems, I wonder.

Macro
Roundtable Fellow
Roundtable Fellow
Posts: 70
Joined: Sat Dec 03, 2011 11:22 am

Post by Macro » Sun Nov 25, 2012 11:00 pm

svquant wrote:From an older Winton pitchbook - perhaps this helps? May have an even more detailed presentation from back when they had less than $500MM AUM. Note as they have grown the marketing message and details have changed greatly but I have seen version of this slide in presentations as late as 2009.

While many CTAs do not talk about moving averages as a forecasting methodology, i.e. just look for crossovers, it is a very well developed field. Just look at any time series forecasting text book and you can see how others look at exponential moving averages for example.

So the mechanisms shown by Winton could all be done via single or double moving averages, i.e. well recognized LTTF techniques.
From that picture, it seems he's employing the Markovitz mean-variance optimization. I fail to see how it is possible to forecast the 'expected returns', 'expected correlations' and 'expected volatility' without succumbing to the hindsight bias.

Edit:

Just read some of the comments by Chris67 and I have to agree.
With regards to volatility forecasting, while it may be perceived that it is easier to forecast market volatility opposed to market returns, conventional vol forecasting methods such as GARCH, ARIMA and VARIMA are also extremely misleading.


Edit 2:
If Winton is using those techniques to allocate across strategies, then it would make sense.

Post Reply