Parameterless Strategies

[O(1)]

Recently I have come across something that I couldn't grasp. In this chat (https://chatwithtraders.com/ep-103-dave-bergstrom/) the speaker talks about parameterless signals to avoid data mining bias and overfitting. However, I am having trouble grasping the concept of parameterless rules... Even a simple "buy if today's open is higher than yesterday's close" rule can be parameterized by adjusting the lookback period for example "buy if today's open is higher than last 2 days' high" , "... last 3 days' high". Can someone elaborate and if possible give an example for such signals/rules?

Maybe he's talking about this..
http://epchan.blogspot.com/2008/05/parameterless-trading-models.html
http://epchan.blogspot.com/2008/08/more-on-parameterless-trading-model.html

I didn't read it thoroughly. But, I'm wondering if he just means allowing variables to be dynamic within the system.. instead of having to manually adjust using user input.

Sounds like mumbo jumbo.. Maybe he's bored?

 

[H2O]

Parameterless signals could indicate a number of separate signals combined using weightings that minimise the correlation between each individual signal. This allows the overall strategy to become 'parameterless' i.e. there are no individual parameters to tweek since the signal is combined thus avoiding overfitting and data mining bias. Essentially, you are fitting to minimise correlation between the individual signals, rather than fitting for performance.

Wouldn't the correlation lookback period be considered a parameter?

 

[runtrader]

Wouldn't the correlation lookback period be considered a parameter?

Yes, you are spot on, the correlation look back period would be a parameter. You've got to start somewhere! The key is to minimise the tweekable parameters.

 

[murray t turtle]

Thanks for the answer.

''SMA cross .......................................................''

As you can see I am very confused, could you give a very basic example of a strategy to clarify?

Yes;Koj ,its not rocket science.
Put it simple, 50 day simple moving average crosses 200 day simple moving average; think about selling -test that. Uptrending 50+ 200 dma may mean a buy.NOT a prediction .Test it with anything you want .IBD may have been among the first to use those measures[ 50 dma , 200dma].; they still do, after all these years........................................................................

Or if you want to waste some more time use MACD, ema, wma, other moving averages. I use some ema..... ; but i'm not confused about it. As they say in chicago-''the smarter you are the longer it takes'' Other things may help but that's a good foundation; i've tested most moving averages including the WSJ 65 dma=, too late, too lagging.LOL