Wednesday 8 October 2008

Derivatives Data Shows No All-Clear for Equities, But is Positive for Precious Metals

They're throwing everything at the credit crunch including the kitchen sink, but the horror show just won't end. That's scary. Here is the way my trading setups see the latest weekly data last Friday from the Commitments of Traders report showing trader positioning in the infamous derivatives markets:

- My S&P 500 trading setup has been in cash for all but one week since early July. It will be giving a bearish signal for a single week during the week of Oct. 13, then goes back to cash. So this setup suggests we're far from an all-clear.

- Large speculators in the Dow Jones industrials - the wrongway traders who are positioned badly at market turns - have built up an extreme bullish position in Dow futures and options, reinforcing my bearish signal in this market. Not a good sign either.

- Commercial hedgers in gold futures and options are massively bullish for the second week in a row - now 1.26 standard deviations above my moving average for this setup. This of course reinforces my bullish signal in that market. Silver commercials have been at similar extremes of bullishness in their net positioning since early August as a percentage of the total open interest.

Portfolio update: I've just updated my portfolio page with results from ongoing and recently closed trades. Not a happy picture, but hey, that's trading. As I've said before, even the best traders win just 60 percent of the time. The key for me is to remain disciplined about following signals and my risk-control rules so I can stay in the market. (More on the latter here.)

TAGS: S&P 500, Dow Jones industrials, silver, gold, COT, Commitments of Traders, market timing, trading system development, CFTC, Commodity Futures Trading Commission, COTs Timer, out-of-sample testing, walk-around testing

8 comments:

Anonymous said...

hi,

a week ago you wrote you won't be trading any signal from the commodities this week because of risk control. so why did you go long gold, then?

johnny

Alex Roslin said...

Hi Johnny,

Thanks for your question. What I said was I wouldn't trade any signals that week because of my risk-control rule. (The rule is that I'll take a new signal only if a majority of highly correlated markets are leaning in the same direction.) As of this Monday, the majority of my highly correlated commodities markets (three of five) are bullish, so I could take the gold bullish signal.

Note that I've taken my heating oil setup offline temporarily while I work on improving its statistical robustness. So that leaves five highly correlated commodities markets. Note also that natural gas is not strongly correlated to those other commodities markets.

Regards,
Alex

Anonymous said...

What's your year to date return using your system?

Alex Roslin said...

Hi Anonymous,

I was losing money YTD in the teens as of Friday's close. The portfolio went into the red in mid-August. I don't disclose the exact numbers mostly for legal reasons. But I don't think the return is a good reflection of the potential of the system or I wouldn't be sticking with it. I've been phasing in better setups through the year and am still doing so, so there's some comparing apples with oranges in that number.

Regards,
Alex

Anonymous said...

slowly getting wiped out, are we? i think a better system in the recent months might have been trading in just the opposite direction to your signals - go short when you get a bullish signal, and go long on a bearish signal. goes to show you that the whole thing is really random and there's no way to predict anything. you were just lucky in the first few months, and not so lucky in the last few. even if your system provides 100% accurate predictions of the past, that still doesn't tell you anything about the future.

Anonymous said...

"But I don't think the return is a good reflection of the potential of the system"

Trading is not an intellectual exercise. If the returns are down, so is the system.....

"or I wouldn't be sticking with it."

That is a statement about your beliefs, not the robustness of the system. Your belief is not proof of system robustness

Alex Roslin said...

Hi Anonymous,

I disagree. Trading can and should be an intellectual enterprise, in my opinion. It certainly shouldn't be an emotional one.

As for the robustness of the system, I've done my best to check on that with the data I've collected. (See the latest signals table.) Using a mechanical system is about taking the "belief" out of the trading.

But even in the best systems I've seen, there are frequent losing trades and frequent periods of declining equity curves. They can be challenging to go through, but that's how it works.

Regards,
Alex

Alex Roslin said...

Hi Anonymous,

Is the market truly random? There's been lots of debate on that, but I think anyone in the market must be there because they believe it's not random. Even a buy-and-holder believes the market generally goes up over the long run.

Regards,
Alex