Ernie Chan claims that statistical arbitrage is better than momentum like trend

HurricaneUS · Aug 9, 2013

Quote from GloriaBrown:

on his new book page 151:

"... I have often found that it is harder to create profitable momentum strategies,....
More...

"I" is the operative word...

GloriaBrown · Aug 9, 2013

It would be wonderful if acrary can see this post and join the discussion.

Rationalize · Aug 10, 2013

Quote from Halal Burger:

Ernie Chan has spent his entire life analysing markets using quantitative techniques and methods, i.e he has been on the academic version of the hunt for the holy grail.

Has he made any money from trading: No.
More...

I suspect you are correct.

While I like his ideas, I think the tracking that he assumes will hold, is not stable enough to run successfully.

Sergio77 · Aug 10, 2013

Quote from sidm:

Yes. I think it will take a lot of money to make it work. Not to mention the real need for automation, hence coding.

For example, suppose the ratio (regression coefficients) for the pair of assets came out to be 1:1.4. Ideally, your mean reverting portfolio should mimic this as closely as possible. Since we can't have fractional units, you would like to have at least 5:7 ratio. That is, the mean reverting portfolio now contains 5 units of the first asset, and 7 units of the second asset. The most convenient instrument to do this type of trading is futures. So you are looking at opening 12 futures contracts at the minimum.

Since there is no guarantees on how far this portfolio will divert from the mean, you need to have a good buffer to avoid margin calls for these 12 contracts.

Also, consider the fact that we are really trading the portfolio (pair) here, not the individual assets. Hence if you want to place some kind of stop-loss, the brokerage won't provide it. Brokerages only provide stop-losses for individual contracts. So you will have to either monitor the pair-portfolio manually or write some code that automates it.

Moreover, don't forget that "surer" the bet (stronger the correlation between two assets), the more crowded that trading space will get. So if you want to do StatArb on ETF vs its components, you better figure out a way to do things faster than everyone else.

StatArb can be rewarding, no doubt. But there is too much work and capital required for smaller players.
More...

Thanks for the examples.

nitro · Aug 10, 2013

Quote from sidm:

Y...Since there is no guarantees on how far this portfolio will divert from the mean, you need to have a good buffer to avoid margin calls for these 12 contracts....
More...

Not if you trade instruments that have some sort of spread margin. In that case, it would be something like two to three "naked' futures, not twelve.

GloriaBrown · Aug 10, 2013

Quote from sidm:

Yes. I think it will take a lot of money to make it work. Not to mention the real need for automation, hence coding.

For example, suppose the ratio (regression coefficients) for the pair of assets came out to be 1:1.4. Ideally, your mean reverting portfolio should mimic this as closely as possible. Since we can't have fractional units, you would like to have at least 5:7 ratio. That is, the mean reverting portfolio now contains 5 units of the first asset, and 7 units of the second asset. The most convenient instrument to do this type of trading is futures. So you are looking at opening 12 futures contracts at the minimum.

Since there is no guarantees on how far this portfolio will divert from the mean, you need to have a good buffer to avoid margin calls for these 12 contracts.

Also, consider the fact that we are really trading the portfolio (pair) here, not the individual assets. Hence if you want to place some kind of stop-loss, the brokerage won't provide it. Brokerages only provide stop-losses for individual contracts. So you will have to either monitor the pair-portfolio manually or write some code that automates it.

Moreover, don't forget that "surer" the bet (stronger the correlation between two assets), the more crowded that trading space will get. So if you want to do StatArb on ETF vs its components, you better figure out a way to do things faster than everyone else.

StatArb can be rewarding, no doubt. But there is too much work and capital required for smaller players.
More...

There are e-micro futures.

Sergio77 · Aug 11, 2013

Ernie Chan is doing institutional trading, big money. This is the reason he is looking at statarb. Retail traders do not qualify other than a few exceptions. There are better blogs for retail traders with very good analysis. Here are two I follow and their methodology is more suitable for the "small guy":

http://quantifiableedges.blogspot.gr/
http://www.priceactionlab.com/Blog/

GloriaBrown · Aug 12, 2013

Quote from Sergio77:

Ernie Chan is doing institutional trading, big money. This is the reason he is looking at statarb. Retail traders do not qualify other than a few exceptions. There are better blogs for retail traders with very good analysis. Here are two I follow and their methodology is more suitable for the "small guy":

http://quantifiableedges.blogspot.gr/
http://www.priceactionlab.com/Blog/
More...

For all Ernie Chan has mentioned, it is possible for retail trader to do it with E-micro futures, it is just the statarb stats part maths are much harder and more steps than average momentum way.

savagemp5 · Oct 7, 2013

Quote from GloriaBrown:

My last post using Ernie Chan terms in the topic making it so confusing so I want to make this new post to talk about what I really mean. Ernie Chan has two books and he claims that statistical arbitrage strategies with two or more derivatives (he means the term Mean-reverting to describe these ) are much better than any kind of "momentum" strategies like trend following.

I don't know how to design a statistical arbitrage strategy to backtest to see which one is better. In his new book, he has like at least 6 math formula to find out a relationship between two derivatives before talking about strategies, and I still don't really get the real formula of these except a general concepts after reading like half of the book.

Except being heavily talking about matlab and very likey matlab sponsor him, he talks very professionally and he had a lot of working experience in statistical arbitrage trading desk in banks and funds. He is not like other "coaches" that just talks without any real prove of making money.

I am one of the active participates in his blog for months. I know if I ask him question like this, he would just reply what he said in his book, so I want to see how other think over here.

For me who doesn't live in US, using statistical arbitrage is something not easy even not consider how to implement because

1. I would need to access to many different derivatives and I have to do both long and short, which means I have to access to US markets since most of these are over there, while my time zone is exactly opposite to US, then I basically either stay awake every night. Even auto trading, I should stay awake to make sure everything goes well.

2. Double commission with at least two derivatives instead of one. In his book there are no commission charge.

3. I need to transfer my money to US dollars, then later transfer back to my own currency. I counted it and this is already at least a 0.2% cost.

Of course if statistical arbitrage is really much more powerful, then all these 3 factors are almost nothing.

So how do you guys think? I am still reading his book and just implement all of his math formula to test out a statistical arbitrage is already much harder than any momentum ways I know.

Please share your thought
More...

If your not living in US = Singapore, pls dont throw my face... The questions you've asked are quite stupid, after going thru the entire thread here.

Here you are asking abt stat arb and there you are wondering about currency risk or comms trade or backtest platform questions. Pls read up on how to hedge currency risk or to get those 30c per trade brokers and backtest platforms.

On your stat arb qns, stat arb is just a general term to strategies that can make money thru statistics, after 300-500 trades everyday for eg. It can be easily done by exploring various asset classes, by playing around with them.

You were asking that u dunno how to design a stat arb for backtesting, what asset class to chose. U are basically like the 95%, just wanting to find out easy ways to make moeny without using "Research" mind of urs. Fyi, there are tons of easy to arbs stocks that are curently being mispriced and in the mean time, im extracting alots of alpha from them . Think LTCM.

The book said 6 months, i say if youhave a good software, research such as backtest on historical prices would easily show u if it is arb-able, how ? Perhaps u need another 5-10 yrs of trading to understand this. Add that and read up more Maths & Quant stuff.

Find out my posts here and on hwz forums this very same nick. TQVM

GloriaBrown · Nov 18, 2013

Quote from comintel:

They both have their role.

Some will prefer one, some another, some both.

It is like asking, which is better, apples or oranges? Cars or bicycles? Stocks or Forex or Futures?
More...

No, these two can be comparable if we only consider these two factors:
statistical arbitrage should be more "safe"
statistical arbitrage has higher winning %

but

momentum like trend may has a home run kind of win sometimes while statistical arbitrage doesn't suppose to surprisingly win big at all in a single trade.