Posted by samspiro on 21st of Oct 2010 at 03:38 pm
I have been testing the mechanical systems for around 60 days
off and on. Tested EOD FAZ, FAS and SRS. Outside of
the last two days, they have had some real ugly numbers. I
think my testing is correct with some small slippage. I
really think the systems are excellent over time and I really want
to start going live but lately, they have not done well in a market
that has gone straight up.
Do they work better in volatility. I was wondering if my
testing is off or if others see the same problem. This
is not meant as a criticism. I really am hoping to implement.
I was looking for some guidance either from Steve or Matt or
anyone else out in the community.
Matt/Steve. Keep up the good work. You are doing a
fantastic job and I want to thank you.
After purchasing the mechanical systems around the same time as
samspiro, I quickly came to the same conclusions. What I have done
instead since then was to develop a few swing trade systems using
Stochastics on 5 min and daily charts. Some of the ETFs (ERY, ERX,
TNA, TZA) works well using daily charts STO(21 periods) with a
small degree of variation on the overbought and oversold levels. I
am using that while waiting for volatility to heat up. Some other
ETFs like SSO, SDS, QLD and ECD works also well on the 5 min charts
but there I am only playing them in one direction, following the
general trend of the market.
I am also convinced that once the VIX is going up, the BPT
systems will be a great asset to have as a trader. I wish that I
had owned these during the Autumn of 2008. However that kind of
period will return one day, we can all be sure about that.
Then coming back to what Matt and Steve have posted many times
about being unbiased to the market actions. Using mechanical
systems have in some weird way helped me to reach that mental state
of mind. It really helps to take out a lot of emotions.
Well it is evening here in Beijing and it is time for the 3rd
beer. Happy trading!
Interesting, Vimal. Could you tell me how you came up with
that VIX number.
BTW, the last week or so some of the eod systems have been doing
well. Specifically, FAS and the AIG systems. And, until
recently, the BIDU system has been performing well. I turned
mine off today because I didn't want to go to Vegas in anticipation
of earnings, but, its profit curve is still looking pretty
good.
I agree - volatility is required for them to work. If volatility
is too low, then you get just enough of a move for them to trigger,
but you get no follow before they reverse on you. As such, I've
found that I have my system off for weeks (and occasionally a few
months) at a time. Mine have been off for a few weeks now, and I'm
glad they are. On a more positive note, since nearly all mechanical
systems are performing terribly, in a fairly consistent manner,
this means that there is an opportunity to create a counter-system
(a mean-reversion type) which will take advantage of these types of
moves. Something like that could be quite profitable. But that
would no doubt stop working as soon as the regular ones start
working again.
That shouldn't be hard to do. All we need to do is to
identify a period in the past with low volatility and then optimize
some systems for that environment. Then we could do a walk
forward test to see how the would have worked recently.
The new SPY daily swing trade system doesn't require volatility,
it works in all markets. bull, bear, sideways, high VIX, low
VIX etc. not released yet though
the system would perform just fine under the leveraged ETF's but
the signals will still only be generated from the SPY. It
will be up to the individual user to decide whether or not they
would like to trade the leveaged ETF's. The average hold
period is around 8 days.
Newsletter
Subscribe to our email list for regular free market updates
as well as a chance to get coupons!
Mechanichal systems
Posted by samspiro on 21st of Oct 2010 at 03:38 pm
I have been testing the mechanical systems for around 60 days off and on. Tested EOD FAZ, FAS and SRS. Outside of the last two days, they have had some real ugly numbers. I think my testing is correct with some small slippage. I really think the systems are excellent over time and I really want to start going live but lately, they have not done well in a market that has gone straight up.
Do they work better in volatility. I was wondering if my testing is off or if others see the same problem. This is not meant as a criticism. I really am hoping to implement. I was looking for some guidance either from Steve or Matt or anyone else out in the community.
Matt/Steve. Keep up the good work. You are doing a fantastic job and I want to thank you.
Mechanical Systems
Posted by etcbari on 22nd of Oct 2010 at 07:36 am
After purchasing the mechanical systems around the same time as samspiro, I quickly came to the same conclusions. What I have done instead since then was to develop a few swing trade systems using Stochastics on 5 min and daily charts. Some of the ETFs (ERY, ERX, TNA, TZA) works well using daily charts STO(21 periods) with a small degree of variation on the overbought and oversold levels. I am using that while waiting for volatility to heat up. Some other ETFs like SSO, SDS, QLD and ECD works also well on the 5 min charts but there I am only playing them in one direction, following the general trend of the market.
I am also convinced that once the VIX is going up, the BPT systems will be a great asset to have as a trader. I wish that I had owned these during the Autumn of 2008. However that kind of period will return one day, we can all be sure about that.
Then coming back to what Matt and Steve have posted many times about being unbiased to the market actions. Using mechanical systems have in some weird way helped me to reach that mental state of mind. It really helps to take out a lot of emotions.
Well it is evening here in Beijing and it is time for the 3rd beer. Happy trading!
Systems are great you are
Posted by vimal on 21st of Oct 2010 at 04:23 pm
Systems are great you are right but from my testing and I am sure Matt/Steve will agree, they need volatility to prosper
My rough calcs say a vol of 25 on the VIX is the minimum required to make them work
Interesting, Vimal. Could you tell
Posted by algyros on 21st of Oct 2010 at 04:35 pm
Interesting, Vimal. Could you tell me how you came up with that VIX number.
BTW, the last week or so some of the eod systems have been doing well. Specifically, FAS and the AIG systems. And, until recently, the BIDU system has been performing well. I turned mine off today because I didn't want to go to Vegas in anticipation of earnings, but, its profit curve is still looking pretty good.
I agree - volatility is
Posted by user32 on 21st of Oct 2010 at 04:28 pm
I agree - volatility is required for them to work. If volatility is too low, then you get just enough of a move for them to trigger, but you get no follow before they reverse on you. As such, I've found that I have my system off for weeks (and occasionally a few months) at a time. Mine have been off for a few weeks now, and I'm glad they are. On a more positive note, since nearly all mechanical systems are performing terribly, in a fairly consistent manner, this means that there is an opportunity to create a counter-system (a mean-reversion type) which will take advantage of these types of moves. Something like that could be quite profitable. But that would no doubt stop working as soon as the regular ones start working again.
That shouldn't be hard to
Posted by algyros on 21st of Oct 2010 at 07:32 pm
That shouldn't be hard to do. All we need to do is to identify a period in the past with low volatility and then optimize some systems for that environment. Then we could do a walk forward test to see how the would have worked recently.
The new SPY daily swing
Posted by matt on 21st of Oct 2010 at 05:21 pm
The new SPY daily swing trade system doesn't require volatility, it works in all markets. bull, bear, sideways, high VIX, low VIX etc. not released yet though
I guess we are all
Posted by dpack on 21st of Oct 2010 at 11:43 pm
I guess we are all waiting for the release!!!
Don't know if you've already
Posted by kbgupta on 21st of Oct 2010 at 06:23 pm
Don't know if you've already mentioned it, but Matt is the SPY system presently long or short? Just curious. tx
neither.......its in cash
Posted by rank10 on 21st of Oct 2010 at 06:43 pm
neither.......its in cash
thanks rank
Posted by kbgupta on 21st of Oct 2010 at 06:45 pm
thanks rank
Are the statistics you posted
Posted by algyros on 21st of Oct 2010 at 05:57 pm
Are the statistics you posted earlier the final ones? And, have you looked into what the system would do if you used leveraged ETFs?
Thanks.
the system would perform just
Posted by rank10 on 21st of Oct 2010 at 06:22 pm
the system would perform just fine under the leveraged ETF's but the signals will still only be generated from the SPY. It will be up to the individual user to decide whether or not they would like to trade the leveaged ETF's. The average hold period is around 8 days.