Question about selling a strategy

[ronblack]

This is from the multicharts wiki
"Historical data available for backtesting will, in most cases, be in the form of time-based bars (minute, daily etc.) based on a group of ticks, with only Open, High, Low, and Close prices available. "

OHLC is worthless, because my strategy needs real data to work correctly.

Do you mean that the logic of your strategy does not use the OHLC of bars but only tick data? If this an HFT system?

There are several backtesters that can do testing at a tick level. You can use R for that also. I do not discount in advance that your system works well. Probably it has some potential but most pros have many requirements for buying a system and these include how it correlates with their other systems, etc. One of my friends works for a hedge fund and they have 4 young guys running all day maybe 10 copies of adaptrade on liquid cooled i7 machines and about 40 copies of Price Action Lab on a private cloud, developing and testing maybe one or two sound strategies per week for commodity futures, forex pairs, ETF and indexes, domestic and international in various timeframes. They also sell systems for markets they do not trade to a few customers in some other hedge funds they work closely together. How can you compete with them? Each system they generate for in-house use or sell to a customer comes with a manual with complete statistics and validation results. I hope you get the idea.

 

[AdrianHagh81]

This is from the multicharts wiki
"Historical data available for backtesting will, in most cases, be in the form of time-based bars (minute, daily etc.) based on a group of ticks, with only Open, High, Low, and Close prices available. "

OHLC is worthless, because my strategy needs real data to work correctly.

That's a bit outdated,
It does support tick by tick analysis now with bar magnifier
provided your data source have historical tick data.
IQfeed provides historical tick data up to 6 months.

PM me if you want any help in backtesting on tick data.

 

[AdrianHagh81]

Do you mean that the logic of your strategy does not use the OHLC of bars but only tick data? If this an HFT system?

There are several backtesters that can do testing at a tick level. You can use R for that also. I do not discount in advance that your system works well. Probably it has some potential but most pros have many requirements for buying a system and these include how it correlates with their other systems, etc. One of my friends works for a hedge fund and they have 4 young guys running all day maybe 10 copies of adaptrade on liquid cooled i7 machines and about 40 copies of Price Action Lab on a private cloud, developing and testing maybe one or two sound strategies per week for commodity futures, forex pairs, ETF and indexes, domestic and international in various timeframes. They also sell systems for markets they do not trade to a few customers in some other hedge funds they work closely together. How can you compete with them? Each system they generate for in-house use or sell to a customer comes with a manual with complete statistics and validation results. I hope you get the idea.

Well, I'm a lone trader and I have SVM and Genetic algos running on 4 HP Blades
connected to a SAN storage 24/7.

And Price action Lab ? for real, I took one look at it and it looks like crappy
VB or pascal code.

 

[TexasTeaTrader]

Do you mean that the logic of your strategy does not use the OHLC of bars but only tick data? If this an HFT system?

There are several backtesters that can do testing at a tick level. You can use R for that also. I do not discount in advance that your system works well. Probably it has some potential but most pros have many requirements for buying a system and these include how it correlates with their other systems, etc. One of my friends works for a hedge fund and they have 4 young guys running all day maybe 10 copies of adaptrade on liquid cooled i7 machines and about 40 copies of Price Action Lab on a private cloud, developing and testing maybe one or two sound strategies per week for commodity futures, forex pairs, ETF and indexes, domestic and international in various timeframes. They also sell systems for markets they do not trade to a few customers in some other hedge funds they work closely together. How can you compete with them? Each system they generate for in-house use or sell to a customer comes with a manual with complete statistics and validation results. I hope you get the idea.

Let me give you a little more detail on it. Its not high frequency but yes it wont run on OHLC of bars, tried it and it only executed 1 trade over a 7 year period. It needs level 1 at the minimum.

This is how it executes a trade: basically it looks for a certain point & figure pattern, when 80% of the pattern is complete it enters a buystop or sell (depending on the direction) at the point where the signal would be 100% valid. If it doesnt fill the order in the current bar and a new bar is formed, then the trade is not executed and cancelled because the signal is no longer valid. Thats why OHLC wont work.

 

[Eyez]

I bid $0.01

 

[NoDoji]

...I know...will continue to work in the future.

That's a bold statement

The margin required to trade a single contract CL and NQ is so minimal with some brokers that I don't see why you don't just save a bit or borrow from a friend/family member the small amount needed to open an account with one of these brokers and run your system.

Unless the potential drawdown of your system is significant. In which case I'm not sure too many people would be interested in buying such a system without a solid track record of audited live trading results.

 

[TexasTeaTrader]

I bid $0.01

Based on your last thread http://www.elitetrader.com/vb/showthread.php?t=277294

I doubt you could even afford that.

 

[TexasTeaTrader]

That's a bold statement

The margin required to trade a single contract CL and NQ is so minimal with some brokers that I don't see why you don't just save a bit or borrow from a friend/family member the small amount needed to open an account with one of these brokers and run your system.

Unless the potential drawdown of your system is significant. In which case I'm not sure too many people would be interested in buying such a system without a solid track record of audited live trading results.

Yeah that's what Im going to do, doesn't look like i could sell it without showing 10 years of OLHC backtesting data

 

[kut2k2]

Do you mean that the logic of your strategy does not use the OHLC of bars but only tick data? If this an HFT system?

There are several backtesters that can do testing at a tick level. You can use R for that also. I do not discount in advance that your system works well. Probably it has some potential but most pros have many requirements for buying a system and these include how it correlates with their other systems, etc. One of my friends works for a hedge fund and they have 4 young guys running all day maybe 10 copies of adaptrade on liquid cooled i7 machines and about 40 copies of Price Action Lab on a private cloud, developing and testing maybe one or two sound strategies per week for commodity futures, forex pairs, ETF and indexes, domestic and international in various timeframes. They also sell systems for markets they do not trade to a few customers in some other hedge funds they work closely together. How can you compete with them? Each system they generate for in-house use or sell to a customer comes with a manual with complete statistics and validation results. I hope you get the idea.

Ron, I assure you that systems turned out by this mass-produced method (especially the ones they reject for themselves) are for shit. For a well-crafted homebrewed system, it's like comparing MREs to a gourmet dinner.

 

[Eyez]

Based on your last thread http://www.elitetrader.com/vb/showthread.php?t=277294

I doubt you could even afford that.

Haha yea, ET'ers all about the easy money / do no work / show no value and demand to be paid. The individual who busted their ass for 3-months is now trading full-time. Go figure.

If you're selling a system, chances are, it probably sucks. (just like the 'news app' which is just a RSS reader, and your forexmonster domain name). No one in their right mind sells a system that works consistently. I personally know traders who walked away from $MM deals because the firm wanted their intellectual property.

Selling a system = red flag