The QID position came within a nickel of stopping out on Friday. It is marked-to-market as of the close and within .5% of stopping out.
The system has selected five diETFs to enter, but the rules will allow only 2 trades on any one day. The other ETFs were TWM, MZZ, and EFU. I would not trade EFU due to the awful slippage I have experienced in the name. MZZ is getting more liquid, but is not nearly as liquid as the other diETFs. As the system trades the open, liquidity is a huge concern.
With QID, SDS, and DXD, the Bamboo will be almost all in, and will have all major indices shorted. Another 4% up on the SPY will take it over $79.00 and leave it under $80.00. The SPX 790-800 levels, (SPY $79-$80) looks like strong resistance, but these diETFs may stop out before resistance puts a halt on the recent advancement.
Since November 15, 2007, the Bamboo has closed 90 trades with 78% of those being winners.
Curious, what does the ‘di’ in diETF stand for? I thought at first it meant Direxion, but SDS/QID are Powershares…
double inverse
Wood, do you use the 3x ETFs in the BB or only the 2x?
Shaz, there was a iBC regular that called them diETFs one time and the name just stuck. I do not remember who it was. CA is correct.
CA, not using the 3x yet. I want more data on them. I will probably wait until there is 1 years worth.
TWM and QID are the best
I’m with you bamboo i’m going heavy short.
I’ll be scaling in shorts tomorrow morning. Technicals look good.
Hey Wood, my system is also calling for those ETFs. What about the correlation inherent in those ETFs? I’ve been discretionarily turning some of them down due to the fact that they’re all almost one big trade, since they move in unison. If you win, you win big. If you lose, you lose big. Any thoughts?
Junk, that is something I’ve been thinking about a lot. There are a couple of ways I’m dealing with it.
1. Divide account into a number of parts, each with their own equity. These do not have to be actual separate accounts.
2. Each part is traded by a different system/setup/strategy
3. Each system/setup/strategy is limited to a set number of trades in one time. For the Big Bamboo, its account can only use 50% of available equity in one day.
What this does is insure that your total equity is not correlated as it is unlikely that all possible systems/setups/strategies are giving the same signal at the same time. Within each segmented account, because you are limiting the amount of exposure allowed for each day, you are able to continue to increase exposure IF the conditions that created the initial entry continue the next day or days, i.e. the edge still exists but may have grown even better.
To sum it all up- Lets assume I have a 100K account split into 4 parts which I trade 4 systems. The Big Bamboo has 25K to trade with. Over the past week it has slowly been getting more and more short. It had 1 position, 5 more triggered, but it will only take 2 in one day. Now, the Big Bamboo is using 75% of available capital (3 out of 4 possible slots are used). If on Monday the conditions are still ripe, it will be able to add on Tuesday the final 4th slot, likely at a better price than on Monday. So now, in essence the Big Bamboo has me about 18% short, even though all the vehicles are correlated (QID, DXD, SDS).
Lets not forget about the other 3 systems. 1 of those systems might be an aggressive mean-reversion system. It may be allowed to go all in (25% of total equity) in one day. Lets pretend it is allowed to go all in. It may have triggered a signal to short the SPY on Friday’s open. So now it is all in short, but total account short exposure is only 25%+18%=43% short, granted they are all correlated.
So while we are now 43% short, the positions were added slowly over the course of a week, not at one day, at one price level.
The other 2 systems may be long only systems, common stock system, whatever. So they may be in cash, long, or whatever.
This is how I handle the correlations. I hope it helps. It makes sense to me but I may not have explained it well.
Down the road my quiver of systems will be diversified enough in structure and time horizon so that correlation becomes even less of an issue.
Wood, i have noticed that some of those diETFs are affected by the magnitude of the move of the underlying. For example, a 2% move in IYR may result in 4% in SRS, but a 10% move in IYR results in 27% move in SRS. this implies that if we had the 4% move in SPY in a single day you may get stopped out, but 4 days of 1% moves may keep you in (referencing the 79/80 level).
As for scaling in. I did a bunch of back-testing on various strategies. My view is that scaling-in reduces draw down but also reduces return over the long run. i.e. if you can stomach the draw downs, and can withstand a long streak of bad trades, it is best on the long run to go all in every trade.
Cool stuff man, I bought some SDS Friday. Does the BB use etfs only related to equities or does it also use etfs related to commodities or currencies?
Born2- I have noticed the same thing, about the magnitude of the moves.
I agree about scaling in. Problem is that if the drawdowns are too harsh, and the psychological issues created by the swings of the system are too uncomfortable, eventually the system will be over-ridden by the user.
Keeping in mind that once one decides to trade a system that he/she must stick with EVERY trade, I try to design them where returns may not be the absolute highest possible but the system doesn’t have extreme drawdowns that would result in someone quitting the system or not taking a trade.
VCU, it does trade dig/dug, but I do not believe the portfolio has been updated to include currencies. A few months ago they were not meeting the minimum volume requirement. They may be now and so I might need to update the portfolio. Honestly I haven’t tested the system over currencies. That is a good idea.
QID did well today. Were you able to stay in?
Nah Gapping it stopped out!