Traders' Roundtable

Thought I would start a thread which lists and perhaps discusses articles relating to our favorite activity... I am sure a lot of the articles listed, including the one I am going to post a link to, will be familiar reading to some, but for newbies like myself, a thread like this should be useful.

http://trendfollowing.com/whitepaper/CMT-Simple.pdf

Interview with Jim Simons from 2000.

http://chinese-school.netfirms.com/abac ... imons.html

Article on Nassim Taleb from the New Yorker, April 22 & 29, 2002.

http://www.gladwell.com/pdf/blowingup.pdf

adamant wrote:Interview with Jim Simons from 2000.

http://chinese-school.netfirms.com/abac ... imons.html

Thanks for sharing this link - I hadn't seen that one. Anyone contemplating the idea of winning at trading would do well to read this article to know what we are all up against.

"Fooling Some of the People All of the Time: The Inefficient Performance and Persistence of Commodity Trading Advisors:"

http://ssrn.com/abstract=1279594

"Trend Following:
Performance, Risk and Correlation Characteristics"


http://trendfollowing.com/whitepaper/tr ... lowing.pdf

http://www.fooledbyrandomness.com/bleedblowup.pdf

Why do people prefer assymetric payoffs?

And now for a little Taleb-bashing:

http://www.efalken.com/papers/Taleb2.html

Stig Ostgaard on the origins of trend following:

http://www.michaelcovel.com/pdfs/stig-ostgaard.pdf

Keep 'em coming Adamant - I like your choices. May I add this old chestnut from Yale Uni:-

http://papers.ssrn.com/sol3/papers.cfm? ... _id=560042

Jim Rogers was quoting this study for awhile. I'm reminded of it everytime I try to trade equities and fail miserably...I find futures much safer.

On the predictive power of moving averages:

http://papers.ssrn.com/sol3/papers.cfm? ... abs=742386

"An Empirical Evaluation of Non-Linear Trading Rules"

http://papers.ssrn.com/sol3/papers.cfm? ... _id=286471

An empirical study of serial correlation in stock returns

http://bora.nhh.no/bitstream/2330/1979/ ... 202008.pdf

A New Anomaly: The Cross-Sectional Profitability of Technical Analysis

http://papers.ssrn.com/sol3/papers.cfm? ... id=1656460

adamant wrote:A New Anomaly: The Cross-Sectional Profitability of Technical Analysis

http://papers.ssrn.com/sol3/papers.cfm? ... id=1656460

"New" --- ROFLMAOPIMP!

That particular "anomaly" has been around since rocks were soft! Plus, the word "anomaly" is about as anachronistic as it gets.

Note those are critiques of the title only, not of the paper itself.

I very much like that "anomaly" paper...

interesting that the 9th decile is often better than the 10th. I noticed something similar when doing ranking on futures or currency returns- switching into the hottest thing was counterproductive- the "next hottest thing" usually then overtook, at which point......you see what i'm getting at.

i always find it strange when authors compare MA returns to Momentum returns though. surely they can see that a 20 period MA(lagged 1 period) descicion rule is the same as the average of all momentum rules 1 through 20.
thus a 20 period MA is probably better compared to a 10-11 period mom, than say a 20 period one. might find better correlation that way.

Re: 9th Vs 10th, remember that trending and momentum aren't the same thing - far from it!

Currencies are a fine example.

Yes, currencies tend to trend, because the economic variables that drive their movements tend to be stable over long periods of time. I don't personally need to know what the %#%$ drives these movements, I only need to observe THAT they are being driven, in order to capitalize on them. However, at least one of the drivers is evident in the technical analysis data itself and I think it pays slightly better to use it.

However, currencies are mean-reverting at *extremes* in momentum. This is because powerful central bankers generally have political incentives to avoid too-rapid moves in their relative currencies and thus, there is reversion.

While there's a lot of similarity between moving-average (and ratio of 2-moving averages) rules and momentum (a.k.a. price change from time x to time y) rules, they are not exactly equivalent. You could spreadsheet this out and see it if you wanted.

ah, i didn't say mom rule 1/2x was EQUIVALENT to MA rule x. I merely stated they are better correlated than mom rule x with ma rule x is(which is what these and other authors often use as comparisons). "You could spreadsheet this out and see it if you wanted."

one way of looking at MAs is as different shapes of collective momentum weighting. exponential,simple,triangular smoothed etc- they are all just basic kernels. momentum rules are equivalent to offset "single period moving averages". some MA's can be looked at as momentum rules with before and after smoothing to take out the jaggediness of raw single momentum lags....

anyway i digress .

i concur that the vol weighting isn't the same as "strength ranking". my point was that it was interesting that an effect i observed in strength ranking(mean reversion of extremes) also is manifested in some minor way in these vol rankings that the authors construct.

Some solid fundamentals for the newer systems trader.

http://www.relativitytradingsystem.com/?page_id=221

Moderator's note: Above link is the blog of system vendor Dean Paul Hoffman, registered with NFA as president of DH Trading Systems LLC