After being out of the market for 5 years, I made a few small stock trades this week. I was shaken out of a nice swing setup earlier in the week. I missed the train on this rally and ended the week even. Hey Mav. I've been using good old pullback, and narrow range patterns. I'm thinking of picking up "The Logical Trader" just to see if it stimulates any creativity. Have you experimented with correlated pairs at all?
It depends what you mean by that. All the risk assets are pretty correlated to each other in a general sense. I don't play the mean reversion game if that is what you are asking. I do spread risk assets against each other. I would like to start analyzing the various risk assets spreads more in depth and possibly look into ACD spread levels.
I have given dozens of trades over the course of this thread. An obvious one would be long SPY/short XLF. You would put the trade on once the SPY confirms a monthly A up or once the XLF confirms a monthly A down. Other trade ideas such as long SPY/short JJC, the copper ETF. Or long USO (oil ETF), short XLF. Or even two long such as long AAPL/long TLT (bond ETF). The long TLT acts as a surrogate short. The possibilities are endless. Probably over 10k possible combinations. I have no interest trading the mean reversion. So I want to actually express a directional view in the form of the spread.
http://www.elitetrader.com/vb/showthread.php?s=&threadid=170318&perpage=6&pagenumber=369 Here is the spread chart I posted earlier of SPY/XLF. Notice how much smoother it is then simply being long SPY all year.
I mentioned before that going forward I'm going to try to point out these spreads as they come up and I'll attach the charts for them for all to follow. What I am simply doing here with these is using what I call an anchor on the spread. The anchor is the ACD signal. In a perfect world, both markets would give you signals but that is not going to happen often with correlated assets. So you choose the anchor and then pick a lagging product to spread it against. We are not playing for mean reversion here, we actually want exposure in the direction of the ACD signal.