mark brown said he found the holy grail.
- Thread starter trend2009
- Start date
I use Tickquest's Neoticker to auto-trade with esignal as a data feed - really like it except it only works with a few brokers....Interactive Brokers, MT Trading are two of them but there is more.
I first heard of Mark Brown in relation to his odd ball system that did really well in the early 2000s. He offered the logic free to the public. Later Lawrence Chan the principle of Tickquest developed a variation of the oddball system of which he also made public.
For me I have a lot of respect for Mark Brown since he offered some of his ideas and insights to the public and that has benefited my own system developement.
I first heard of Mark Brown in relation to his odd ball system that did really well in the early 2000s. He offered the logic free to the public. Later Lawrence Chan the principle of Tickquest developed a variation of the oddball system of which he also made public.
For me I have a lot of respect for Mark Brown since he offered some of his ideas and insights to the public and that has benefited my own system developement.
Well, having only read this far, you only want to use Multicharts with either IB or PFG Best. Multicharts is for advanced technical analysis and made for easylanguage. Those two brokers should definitely be on your list, but I suggest IB if you're an advisor in need of RIA or Brokerage Custodian assistance and PFG Best by far for all things futures.
Please mention Beau Wolinsky referred you to all of them.
we didn't have any luck with any of the trading platforms out there because they were all geared towards retail operations. so we broke down and did our own. i really can't suggest any software that i know of for retail automation. we keep developing tradevec meanwhile.
i will say that i would think that tick quest would be a good start for someone, and heard that multi charts is coming along. i know the owners of both these products.
m
i will say that i would think that tick quest would be a good start for someone, and heard that multi charts is coming along. i know the owners of both these products.
m
Mark,
I didn't quite realize you were posting on this thread...I have since read through it. As such I want to take the opportunity to thank you for this info and really thank you again as the info you have provided to the public through the years is really the cornerstone for any successful developement I have accomplished.
One question I come to is in regard to how one determines that a system works into the future. i.e. you've done your development including walk forward testing...and you begin to trade your account - I am a 1 lot futures trader -
How do you decide if and when your strategy no longer works.
I have been using 1.5 to 2 times historical drawdown as a measure. When drawdown exceeds that amount I figure markets have changed or I curve fit or something and I stop trading. Hopefully some of the drawdown is on paper and not real money i.e. I sometimes wait for a system drawdown before start trading. I usually write out a trading plan with these items written down and what actions I will take.
For instance I was trading a system I developed and it had 4 losing months in a row I waited till after the first month to begin with real money...and after the next 3 months it hit it's drawdown limit (last January) and I stopped trading...since then on paper it came back and for the year is up significantly. Btw -the historical track record was quite good.
I sometimes think if I reach max drawdown I would start trading real money again if it reaches new equity highs on paper or some such strategy.
Actually many of these ideas I got from reading Larry William's books.
I think I also asked this question of Lawrence Chan sometime back and he suggested that you could see the trading statistics deteriorate well before the system blew up but to be honest I didn't fully understand what he meant.
I didn't quite realize you were posting on this thread...I have since read through it. As such I want to take the opportunity to thank you for this info and really thank you again as the info you have provided to the public through the years is really the cornerstone for any successful developement I have accomplished.
One question I come to is in regard to how one determines that a system works into the future. i.e. you've done your development including walk forward testing...and you begin to trade your account - I am a 1 lot futures trader -
How do you decide if and when your strategy no longer works.
I have been using 1.5 to 2 times historical drawdown as a measure. When drawdown exceeds that amount I figure markets have changed or I curve fit or something and I stop trading. Hopefully some of the drawdown is on paper and not real money i.e. I sometimes wait for a system drawdown before start trading. I usually write out a trading plan with these items written down and what actions I will take.
For instance I was trading a system I developed and it had 4 losing months in a row I waited till after the first month to begin with real money...and after the next 3 months it hit it's drawdown limit (last January) and I stopped trading...since then on paper it came back and for the year is up significantly. Btw -the historical track record was quite good.
I sometimes think if I reach max drawdown I would start trading real money again if it reaches new equity highs on paper or some such strategy.
Actually many of these ideas I got from reading Larry William's books.
I think I also asked this question of Lawrence Chan sometime back and he suggested that you could see the trading statistics deteriorate well before the system blew up but to be honest I didn't fully understand what he meant.
Mark your latest comment was a very broad generalization. I suspect it is related to some sort of mean reverting scenario. It might have some relevance over the course of a day or a week.
The prior post described a situation that occurred over a four month period. Market dynamics need to be reaccessed relative to the assumptions of the trading model and deviations from the back tested period.
With four months of trading data one should be able to establish the source of actual vs. expected deviation.
It is unusual that after 4 months the market reverted back to true track. Without additional information on the market, time period traded, and system used, additional insight can not be provided.
The prior post described a situation that occurred over a four month period. Market dynamics need to be reaccessed relative to the assumptions of the trading model and deviations from the back tested period.
With four months of trading data one should be able to establish the source of actual vs. expected deviation.
It is unusual that after 4 months the market reverted back to true track. Without additional information on the market, time period traded, and system used, additional insight can not be provided.
the big money pools who get in and out less frequently are a good reason to be aware of larger time frame scenarios,for example(not fact) the initial idea to use QE on a grand scale allowed the banks to profit enormously,since they were the ones working with bernanke,they knew a few weeks,a month ahead of time that more or none was coming and could position themselves accordingly,these changes in allocation may or may not show up on a 4 month read