WANTED: Stock Tickers to Back-Test a New Price Swing Indicator

[AnalystM]

I need you experienced stock traders to post back with stocks that you are very familiar with so I can back-test a new indicator I'm developing for a price-swing buy-in and exit strategy. Stock suggestions should have the following characteristics:

1) traded on a major exchange for over 5 years.
2) preferably tend to exhibit ~2 or more +/- price swings per year (e.g., in the last 3 years) of 20%+ each way.
3) SINGLE COMPANY stocks only, no ETFs (too much "noise").
4) are NOT currently caught in a prolonged "value trap" (fundamental company issues).
5) are NOT unduly prone to events or news (no emerging biotechs for example).
6) are NOT affected by recent major cliffs or gaps (again too much event-influenced, such as by lawsuits, etc.).

Good examples are: GM, ACM, AAP, AOI.

If I use your stocks in my analysis testing, I will post the results here.

Thanks so much. AnalystM

 

[Quantt]

I need you experienced stock traders to post back with stocks that you are very familiar with so I can back-test a new indicator I'm developing for a price-swing buy-in and exit strategy. Stock suggestions should have the following characteristics:

1) traded on a major exchange for over 5 years.
2) preferably tend to exhibit ~2 or more +/- price swings per year (e.g., in the last 3 years) of 20%+ each way.
3) SINGLE COMPANY stocks only, no ETFs (too much "noise").
4) are NOT currently caught in a prolonged "value trap" (fundamental company issues).
5) are NOT unduly prone to events or news (no emerging biotechs for example).
6) are NOT affected by recent major cliffs or gaps (again too much event-influenced, such as by lawsuits, etc.).

Good examples are: GM, ACM, AAP, AOI.

If I use your stocks in my analysis testing, I will post the results here.

Thanks so much. AnalystM

If you are testing a system, you should test on the entire market or on as much broader universe as possible or you are fall into data fitting trap if selecting only handful of stocks...

 

[AnalystM]

Thanks for the reply. Can you please elaborate on what you refer to as a "data fitting trap" as it relates to one or a small group of stocks as opposed to a "broader universe as possible?" I'd like your perspective on that as a quant or technical trader.

 

[Quantt]

Thanks for the reply. Can you please elaborate on what you refer to as a "data fitting trap" as it relates to one or a small group of stocks as opposed to a "broader universe as possible?" I'd like your perspective on that as a quant or technical trader.

The problem is over fitting, e.g. you are looking into one or few companies and you over fit your strategy to be perfectly optimizing just for them and if something fundamentally changes, the chance is, it wont work anymore... Over fitting can also occur when using limited amounts of data as well, for example optimizing the strategy based on data from the last few years only - e.g. from the last bull market cycle... The best way to build a robust system is to test on as much of stocks as possible during all market cycles...

 

[AnalystM]

Thanks Quantt for the clarification. I totally agree with you that trying to use a few stocks to "fine-tune" a strategy is not correct. But that's not what I'm trying to do. I have a pricing indicator that I want to test against different stocks to gauge its predictive probability, and where it fails, I want to determine what concurrent conditions existed that might have predicted that failure.

I am simply asking for stock tickers to test, from independent investors such as yourself, so I can't accuse myself of cherry-picking my own test examples. Thanks.

 

[Quantt]

Thanks Quantt for the clarification. I totally agree with you that trying to use a few stocks to "fine-tune" a strategy is not correct. But that's not what I'm trying to do. I have a pricing indicator that I want to test against different stocks to gauge its predictive probability, and where it fails, I want to determine what concurrent conditions existed that might have predicted that failure.

I am simply asking for stock tickers to test, from independent investors such as yourself, so I can't accuse myself of cherry-picking my own test examples. Thanks.

OK, cool, unfortunately, I cannot help you with specific stocks, because my scanners work on the whole market, so I don't have a favorite basket of stocks to provide...