Statarb needs a lot of money to work. Momentum trading you can do with $100. As far as the blogger mentioned I was never able to make anything useful out of this posts. Maybe he knows but his seminars are for newbies. I tried once to use his concept of Kelly ratio for estimating leverage and the numbers I was getting were just ludicrous.
I think Kelly ratio only makes sense when we have a very good strategy that works so well or few strategies work so well. What is the point to calculate the Kelly ratio if we don't have a strategy that works well in real world? Even if we calculate Kelly ratio, it only makes sense if the strategy is as good as backtest. Base on what I know, stat arb can work as long as we can buy two things like 2 futures or 1 future + 1 etf. Do you guys think so? The worst thing of stat arb is just like what Mr. Chan mentioned, there is almost impossible to set a stop lose like momentum, then if we don't know if the relationship inside the stat arb strategy is still working, we can lose crazy many money in one trade without even notice it is happening.
I just read this from his new book, page 183: Furthermore, these kinds of "turncoat" price series that regime-change from mean reversion to momentum would never show up in our catalog of profitable mean reversion strategies because our catalog would not have included mean reverting strategies that failed in their backtests. I don't get this at all. So he thinks backtest can always predict the future for mean reversion? As long as backtest is profitable, then that mean reversion relationship will hold forever and be profitable forever? He seem like tries to emphasize mean reversion too much and hide what he mentioned at the end of the book: it is too hard to have risk management with mean reversion, once the predicted relationship between 2 or more derivatives don't work, then it can be a big lose in one wrong trade and too late to cut before any notice.
Please back up with a quote your assertion that Chan claims that stat. arb strategies are much better than any kind of momentum strategies or withdraw your assertion. It seems very unlikely he would make any such ridiculous assertion.
That definition is too narrow. An arbitrage does not have to be between two or more of anything. The arbitrage part is very straight-forward. It is the statistical part that confuses people it seems to automatically mean pairs trading or something like it. The meaing of the word arbitrage does not change because you attach statistical in front of it.
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.
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.
on his new book page 151: "Letâs start with the cons. In my own trading experience, I have often found that it is harder to create profitable momentum strategies, and those that are profitable tend to have lower Sharpe ratios than mean-reversal strategies." This is basically the main concept in his new book. He mentioned and explained this concept throughout his whole new book, you can read by yourself and see many examples like the one above.
I think Ernie Chan is great and his books, courses, blog all are win-win for himself and others. I just don't really like he only gives out matlab code but I can understand this is another way for him to earn some money. I think he is good enough to be individual trader and have pretty good earning comparing to average people , while he shares many good main concepts and points out the correct way to do your own research if you want to be a serious trader. Well he never points out a killer level strategy in his books but I think that is not his goal and not what I expect. He may has few really good strategies but there is no way he should share with public. His new book is the best out of all trading books I have read so far, and I have read like at least 50 books. I track what he does and honestly I don't think he tries very hard for marketing his books or other stuff that he can have earning, so I don't feel he really earns his living just from books and marketing. I really don't think he needs to just depend on marketing his book with his working experience. If he just keeps working in ibanks, I am sure he can earn much better salary than his book. His new book is a complicated one and definitely not interesting for general public. If he wants to earn more, he should write a much easier one with eye catching topic. I feel he just writes what he is good at. I start to thread just to see if you guys also think if statistical arbitrage is really better than momentum or not.