I was wondering if anyone has had any experience with pair-trading or statistical arbitrage. I am interested in it because I have an opportunity to learn from some local traders in my area through an extensive course. In all of the introductory trading books that I have recently read it is not mentioned at all. Nor have seen any statarb books in particular.
The only things I have read about it are the wiki entries and other brief summaries. I was hoping someone had more experience in this. I became very interested when reading about a trader who posts on trade2win (a trading forum with a pretty bad signal:noise ratio, like all forums), grey1 (iraj), who said he liked arbitrage methods and was very successful. His posts were inspiring to say the least. My interest in statarb is not to use institutional money to trade, but just to use my own money.
Thank you in advance for your replies.
Subscribe to:
Post Comments (Atom)
I am VERY GLAD that posted this, because I am also fascinated by arbitrage. After all, trading is all about finding high-probability ways to make money... and that is the definition of arbitrage! Even though many people consider today's markets to be "efficient," we traders know that there are always inefficiencies that can be exploited. Even though there are many arbitragers out there that try to keep things as efficient and synced as possible... they always miss some things. It is that - the rare and temporary inefficiencies in the market - that give an "almost guaranteed" trading opportunity if you can just flip the security and profit off the discrepancy.
ReplyDeleteDo I have experience in arbitrage? Yes, but I am still relatively new to the stage. I currently use an "arbitrage"-based method that uses lagtimes as my primary way to trade. I daytrade NVLS, which is a semiconductor stock. I have noticed that it is strongly correlated to sister stocks like KLAC and AMAT, follows the SMH exchange-traded fund for semiconductors extremely well, and it also follows the Nasdaq and S&P. Essentially, these securities are "linked." The order is simple: S&P leads everything, which is usually lockstep with Nasdaq. These markets then lift ETFs, like SMH. SMH then lifts its holdings, like NVLS. In other words, a rising tide (the market) lifts all boats (the sectors), and the boats lift their passengers (the individual stocks).
That's the basic idea. However, there is a type of chain-reaction that occurs, and it gives me lagtime. I subscribe to an S&P squak box, so I hear the roar of the S&P 500 futures pit (which is the most powerful market leader). If I hear hysteria in the pit and see the ES contract dropping, I have a couple seconds to short NVLS. Ideally, I hear the screams, I confirm the drop in the ES, I short NVLS, I see SMH drop, and then I see NVLS panic to keep up with the sector, and voila! Arbitrage. I profited off a temporary lagtime in the market's ability to sync itself.
This is my basic scalping method. Currently, I am simulating this method in the hopes that once I have over $25k in my account, I will be able to make a living off of this type of play.
A few caveats and one more thought. First of all, you might ask me, "Well what happens when the S&P drops and NVLS pops? You can't be right all of the time." Absolutely. The method won't work all of the time; sometimes it will be dead wrong. I know that. I'm simply playing a numbers game - like a casino - knowing that after 100 trades, I will have made a lot of money, even if I have no guarantee about the results of an individual trade. Second, you might ask me, "Well gee, you just spilled the beans there. If your method works, why did you just tell all of us?" Honestly, I already know that NVLS will one day change its behavior and go into its own world. All I'm doing is trading a temporary method that will only exist for a short amount of time in the marketplace because it is, at the moment, inefficient. Once NVLS stops behaving, I will leave. I'm not emotionally attached to NVLS. I just am familiar with its behavior and how it responds to changes in the market and its sector. There are dozens of other stocks that do the exact same thing, and I am happy to move on to them once NVLS stops performing. The concept, though, is sound: a rising tide lifts most boats and their passengers. That's the concept, and that will always be true.
So, I hope that helps. You are welcome to email me and chat more about this if you want: mikeanthony5000@gmail.com. I would be very interested in YOUR thoughts and anything you have discovered!
One final thought is very complex, but as this is arbitrage, you might expect that :-). I have a friend who uses an interesting method of "guaranteed trading" to increase his profits in pre- and post-market trading. He simply buys below and the inside bid and sells at the inside bid! Now that's guaranteed trading if I ever heard it! He does this by routing his order to famous routes like ARCA and NSDQ in the pre- and post-market when these ECNs are posting a below-the-inside-bid bid. Obviously this doesn't happen too often, but again, this is arbitrage, and the fact is is that it sometimes happens... and arbitragers profit off those temporary discrepancies. Anyway, so he sets up an invisible order a penny above ARCA's and NSDQ's best bid, and then just waits for an idiot to click "sell ARCA bid" or "sell NSDQ bid" which is an option available on some trading platforms. Unfortunately for these traders, they just sold BELOW the inside bid, and my friend picks up their shares with his invisible order! Brilliant. Now he has shares that have guaranteed profit, and he can either sell them immediately at the inside bid, or if he likes the trade, he can even try to hold them for more.
ReplyDeleteHe says that this doesn't happen that often; but then again, he doesn't scan the entire market for this opportunity either. He says that about once a month, he'll get a crazy fill that will put him 10% below the inside bid. He makes about $4k/year just from this practice. Pretty cool. Pretty arbitrage.
So, anyway, those are a sampling of my thoughts, and I have a lot more! I love arbitrage. Hopefully this helped and you feel comfortable emailing me for anything else. Cheers! I wish you the very best in your search for high-probability trading opportunities.
The method I was particularly interested in was a semi-automated daytrade of a pair of stocks. It is even more intriguing to me since it seems like it's not very well known or popular. The thread I was reading described 15 to 30 minute trades when the market was flat, for example, INDU did not go past plus or minus 50 during the opening range.
ReplyDeleteThis person had 50 to 100 pairs he refreshed every few weeks or double checked their correlation. I think he eventually got this fully automated and got his VWAP bands and exhaustion engine coded into his tradestation. He described that we could do something similar with MPD bands. I think the VWAP bands were used for entries and exits.
The trades were scaled in and scaled out and he was very strict about risk/money management. He used to work for an institution where they backtested this method and it was not to be used during trending markets. He then went on his own to trade only his own money and refined his own technique and became VERY successful with it, from what I gathered.
I don't have access to some other of his excellent posts because he was a member of this forum from 04 to 2008 then is now gone and some of the forums need passwords and he's not there to issue them - which he used to do for free. I would like to learn more since I like to know about every type of trading.
Superb! What you described is exactly the type of odd, quirky, temporary inefficiencies that exist in the market. You're on the right track!
ReplyDelete