Got gold?

General discussions about futures.
Post Reply
danZman
Roundtable Knight
Roundtable Knight
Posts: 117
Joined: Thu Mar 10, 2005 10:07 am
Location: Newport Beach,CA

Got gold?

Post by danZman » Fri Dec 04, 2009 5:07 am

Just wondering who's NOT in gold right now.

This is a trend following board (mostly). Any reason not to hold?

D

adamant
Roundtable Knight
Roundtable Knight
Posts: 118
Joined: Mon Aug 25, 2008 1:20 am
Location: Norway

Post by adamant » Fri Dec 04, 2009 8:22 am

None that does not involve funnymental arguments. That being said, I sold all my gold stocks yesterday after a very nice ride. No doubt they will go much higher....

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Fri Dec 04, 2009 9:00 am

More precisely, this is a mechanical system traders forum, as it says at the top of the page. Plenty of mechanical systems operate on principles other than "trend following" -- such as mean reversion, pattern matching, good old support+resistance, and the darling of the Wealth-Lab website, "dip buying".

As for gold, it would be hard to find a "trend following" mechanical system that isn't long gold right about now ... chart below. (The chart shows raw contract prices, i.e., it is neither a continuous contract nor a backadjusted series. It's what CSI calls "Nth Nearest" using N=1.)
Attachments
gold_futures.png
NYMEX Gold futures
gold_futures.png (33.97 KiB) Viewed 6488 times

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Post by LeviF » Fri Dec 04, 2009 9:12 am

Last night my system signaled it was time to take profits on a portion of my gold position.

rajivm
Roundtable Knight
Roundtable Knight
Posts: 129
Joined: Fri Jul 25, 2003 12:50 am
Location: New Delhi, India

Re: Got gold?

Post by rajivm » Fri Dec 04, 2009 10:48 am

danZman wrote:Just wondering who's NOT in gold right now.

This is a trend following board (mostly). Any reason not to hold?

D
I guess Last Man already boarded the train.......
Better run for cover now.

danZman
Roundtable Knight
Roundtable Knight
Posts: 117
Joined: Thu Mar 10, 2005 10:07 am
Location: Newport Beach,CA

Post by danZman » Fri Dec 04, 2009 1:52 pm

I could only find one mechanical model that is short gold...it uses the COT data and a time delay.

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Post by LeviF » Fri Dec 04, 2009 2:05 pm

Today would have been one h*ll of a day to be short.

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Post by LeviF » Fri Dec 04, 2009 2:07 pm

sluggo wrote:Plenty of mechanical systems operate on principles other than "trend following" -- such as mean reversion, pattern matching, good old support+resistance, and the darling of the Wealth-Lab website, "dip buying".
Anybody here trade any of these types of systems [profitably]?

nodoodahs
Roundtable Knight
Roundtable Knight
Posts: 218
Joined: Wed Aug 09, 2006 4:01 pm

Post by nodoodahs » Fri Dec 04, 2009 5:14 pm

Not in gold. In emerging market stocks. Momo+trendfollowing system. Gold is not in the top ranking by my method (but certainly is in a "trend" by my method)

adamant
Roundtable Knight
Roundtable Knight
Posts: 118
Joined: Mon Aug 25, 2008 1:20 am
Location: Norway

Post by adamant » Fri Dec 04, 2009 5:56 pm

Feeling very lucky at the moment...gold took a beating today.

That said, the trend is up and I will probably be in again soon enough.

More or less related question: I am experimenting with the Mama Fama as my primary trend indicator/filter using both weekly and daily input and fine-tuning with a couple of classic secondary indicators for entries (not in trading blox as I can't program for s**t, unfortunately). Seems to work reasonably well and be worthy of some more research. But has anyone found the MF crossover to be distinctly better than EMAs to define trend? The research I have found so far seems to indicate that it is only marginally better than EMAS (S&C article).

MAMA - FAMA article attached.
Attachments
mama_fama.pdf
MAMA - FAMA, warning: contains maths!
(80.28 KiB) Downloaded 1019 times

Asamat
Roundtable Knight
Roundtable Knight
Posts: 175
Joined: Fri Jun 03, 2005 7:50 am
Location: Walldorf, Germany

Post by Asamat » Sat Dec 05, 2009 8:54 am

I never understood why people bother to distinguish between different types of moving averages, in the context of building systems which should be stable over decades. If the system is robust and uses one of them, it should work with any of them. If it only works with one and not the other, I would be suspicious. In this case I would try to find out, what this MA actually measures/captures, because it can't be what all the other MAs measure/capture.

The only differences I see are technical points, like ease of calculation, priming requirements, performance.

nodoodahs
Roundtable Knight
Roundtable Knight
Posts: 218
Joined: Wed Aug 09, 2006 4:01 pm

Post by nodoodahs » Sat Dec 05, 2009 9:11 am

There are a couple of theoretical differences.

Often I prefer EMA to SMA because the SMA counts every data point TWICE - once when it enters the average and once when it leaves - so the EMA presents a smoother graph and potentially fewer whipsaws.

However, when I was adapting a "time stop" system from different periods, i.e. check for these conditions once a month, make moves, check back next month, moving that to be able to check daily - I found that using the SMA more accurately reflected the performance (which issues I was in, turnover frequency, etc.) of the longer-period system on my data.

That said, using different types of MA as well as different lengths can also be a good check for the robustness of the parameters!

For example, if my system optimizes at 50 SMA and 200 SMA parameters, I should check different short and long averages around the range in both types, i.e. 40-60 SMA and EMA and 160-240 SMA and EMA. If the system is robust I should get a smooth, gradually changing result around my optimum choices. If the system isn't robust, then small changes from optimum would mean big changes in results.

Post Reply