Equis & Reuters Liable for this?

D

dave4532

I used Metastock v7 series for several years to develop and backtest trading systems from mid 1999 to mid 2003 but then I gave up and I got a regular job because I could not get any real trading results to match my backtest results. All of my systems were based on chart patterns combined with SMAvgs for short term trend confirmation.

Recently I read a paper, which demonstrated some very severe defects in the backtester of two popular programs. These defects caused a number of trades not to be included in backtest results and also caused other trades to have a delayed entry or no entry at all. I was told in another forum that one of these programs in the paper is Metastock 7.0 and I was shocked, boy was I shocked.

Thinking back to what I was doing then, the defects shown in the paper could have really screwed my backtesting results and as a result real-time trading performance could substantially deviate from them.

I wonder if Equis has responded to these claims and whether there are some liability issues here on the basis of a defective product. I am thinking of contacting a lawyer for the possibility of a class A Action after I manage to get the full detailed report from the author. If there are any members of this forum who used older versions of Metastock in the past please send a private message and maybe we get organized.

Link to paper: http://www.tradingpatterns.com/Backtesting.pdf
 
You're not serious!

One of the 1st things a user acknowledges in using most software is not to sue the firm that produced it.

Are you the one who sued McD for that extra hot cup of coffee?
 
Quote from dave4532:

Recently I read a paper, which demonstrated some very severe defects in the backtester of two popular programs. These defects caused a number of trades not to be included in backtest results and also caused other trades to have a delayed entry or no entry at all. I was told in another forum that one of these programs in the paper is Metastock 7.0 and I was shocked, boy was I shocked.

You were SHOCKED because someone TOLD you that Metastock might be one of those programs??? LOL.

Thinking back to what I was doing then, the defects shown in the paper could have really screwed my backtesting results and as a result real-time trading performance could substantially deviate from them. I wonder if Equis has responded to these claims and whether there are some liability issues here on the basis of a defective product. I am thinking of contacting a lawyer for the possibility of a class A Action after I manage to get the full detailed report from the author. If there are any members of this forum who used older versions of Metastock in the past please send a private message and maybe we get organized.

Even if Metastock had this defect, how do you know that it was in your version and even if it was, how do you know that it wasn't corrected in an update or subsequent new version?
More...
Realistically, check the product's limited warranty. All they guarantee is replacement of defective disks and at their discretion, refund of the purchase price. They have NO responsibility for lost profits, bad investment decisions, or for that matter, anything else. Class action suit? LOLOLOL
 
No, I do not also think they can get anything for bad investment/development decisions but it all depends on local State Law and case law. If I remember from past other cases that it depends on whether the manufacturere knew of the defect and purposly did not fix it by recalling the product. I wouldn't mind getting back what I paid if the product was defective. If the class is large enough this may cost to the company a leg and an arm. They probably deserve it for releasing software to the public without paying an expert to test it thoroughly.
 
Quote from dave4532:
I was shocked, boy was I shocked.
More...
I downloaded this paper before and it struck me as pretty bad. The author talks about "high frequency" trading and using "bars" in the same paper -- are they joking? If HFT firms used "bars" rather than tick data, my guess is they'd be out of business literally within a few hours.

The fact of the matter is, no backtester is going to be 100% accurate, whether manual or automated. This is something you're going to have to live with; if you can't, you probably shouldn't be using one. (And, for that matter, you probably shouldn't be trading, either, as this is a field virtually defined by uncertainty.)
 
Quote from Occam:

I downloaded this paper before and it struck me as pretty bad. The author talks about "high frequency" trading and using "bars" in the same paper -- are they joking? If HFT firms used "bars" rather than tick data, my guess is they'd be out of business literally within a few hours.

The fact of the matter is, no backtester is going to be 100% accurate, whether manual or automated. This is something you're going to have to live with; if you can't, you probably shouldn't be using one. (And, for that matter, you probably shouldn't be trading, either, as this is a field virtually defined by uncertainty.)
More...
It would appear that the company that wrote that ... ahem ... "paper" is the one out of business...

http://www.tradingpatterns.com/
 
Quote from Occam:

I downloaded this paper before and it struck me as pretty bad. The author talks about "high frequency" trading and using "bars" in the same paper -- are they joking? If HFT firms used "bars" rather than tick data, my guess is they'd be out of business literally within a few hours.
More...

I read this paper carefully, it took me all morning. IMO it is a well presented analysis. The author is not talking about HFT but at one point he says:

"The effects of the hidden conditions in PA backtesting on high-frequency trading strategies may not be immediately realized unless one performs a detailed and careful analysis of the results but they are significant."

He obsiously means high frequency intraday trading timeframes based on minute intervals or even sub-minute. This is as opposed to swing or position trading. Also, his examples are very clear about what he is talking about..

Quote from Occam:

The fact of the matter is, no backtester is going to be 100% accurate, whether manual or automated. This is something you're going to have to live with; if you can't, you probably shouldn't be using one. (And, for that matter, you probably shouldn't be trading, either, as this is a field virtually defined by uncertainty.)
More...

I don't think you understand what you are talking about. You have 10 backtesters out there and one or two produce different results for the same benchmark system. They are obviously defective. If a backtester missed signals it should have otherwise generated, then it is defective. If you like to use defective products, it is your own right and responsibility. You are not welcome to try to seduce people into accepting such defects.

The paper was excellent I think, well presented and shocking to a degree. I contacted the author for the full paper and if he will not sent it to me, my lawyer will try to get it.
 
Quote from alexandermerwe:

If a backtester missed signals it should have otherwise generated, then it is defective.
More...

I agree. Those who don't agree, they do so for a reason they know. By the way, the website says the full paper if available now. I will try to get a copy of it

http://www.tradingpatterns.com/
 
Quote from alexandermerwe:

I read this paper carefully, it took me all morning. IMO it is a well presented analysis. The author is not talking about HFT but at one point he says:

"The effects of the hidden conditions in PA backtesting on high-frequency trading strategies may not be immediately realized unless one performs a detailed and careful analysis of the results but they are significant."

He obsiously means high frequency intraday trading timeframes based on minute intervals or even sub-minute. This is as opposed to swing or position trading. Also, his examples are very clear about what he is talking about..



I don't think you understand what you are talking about. You have 10 backtesters out there and one or two produce different results for the same benchmark system. They are obviously defective. If a backtester missed signals it should have otherwise generated, then it is defective. If you like to use defective products, it is your own right and responsibility. You are not welcome to try to seduce people into accepting such defects.

The paper was excellent I think, well presented and shocking to a degree. I contacted the author for the full paper and if he will not sent it to me, my lawyer will try to get it.
More...

And MH (the writer) never made a dime trading....lol
 
You must log in or register to reply here.
Back
Top