The reason you don't skip is called market risk. Assume the market goes up or down in a big move and you are on the wrong side of the trade, you are going to get hurt.
To avoid this problem, one can day trade after reports are out speculating on market direction.
Assuming you have an edge, you will win more than you lose.
Problem then becomes psychological in making sure you take your stops and letting your winners run to their targets.
To avoid this problem, one can day trade after reports are out speculating on market direction.
Assuming you have an edge, you will win more than you lose.
Problem then becomes psychological in making sure you take your stops and letting your winners run to their targets.
Quote from bwolinsky:
QLD Projection: 45.9908338693472 QLD Close: 49.5499992370605
QLD Projection: 48.6352731184526 QLD Close: 49.5499992370605
QLD Projection: 47.7039049421038 QLD Close: 49.5499992370605
QID Needs to drop -0.90856233740546%
Up Threshold is $ 23.5342164448662 on QID
It would appear we are settling in on the highs of the year. Given the slope I think it is reasonable for the second projection to go up further from here, so maybe 49.6 isn't overvalued after all. QID closed at 23.75, which is about .19 less than I paid to exit my trade from 23.94 to 24.3. Nice out, good positive slippage, especially on low twenty dollar stocks. A penny per share is a lot bigger percentage wise than on the QLD above $45.
Realistically, though, if you have to worry about 2 or 3 cents each way on your system, I don't really think that system works, but that's just me. I could care less if I had even a nickel of slippage, because the average profits are still way more than tenths of a percent.
That really all goes back to the first post on the previous page. All I'm really saying is that given that you get a certain price at a certain ask or bid, a few cents either way won't affect your results materially. If you disagree then you have a problem, because you don't know what you're doing.
If you're looking for more information, you can see the thread about stat arb pair trading in US Equities.
http://elitetrader.com/vb/newreply.php?s=&action=newreply&postid=2572894
Indeed, almost every single person I've heard explain the pair trade missed the first key step, which is to find the correlation, but once you've done that, you actually peeked to do it, which is a form of look ahead bias. Then you have some talking heads on the street go out and say "Well, I'm short the XLF, but long BAC" when these correlations aren't strong enough to produce a profitable pairs trade.
The best pairs trades are with negatively correlated pairs. If you've ever charted what a negatively correlated pair looks like and compared the volatility of the ratio with a positively correlated pair, you'd notice the patterns are quite choppy, which serves to reduce the effectiveness of whatever your overbought or oversold level is. The last point I would make is the fallacy that you must short a security and be long simultaneously. In the case of negatively correlated pairs, you can just buy the undervalued and get the same result without extra commission. For positively correlated pairs, the approach works just as well. All you do is buy the undervalued, and skip shorting the overvalued because you know the correlation will hold and you can save your commission money for another trade.