What kind of strategy would you call this?

P

pistolpt

I've enclosed a chart which holds one of the several keys to my investment strategy. This is based off of about 2,000 days of observations.
The prior day's value of a certain statistic is on the x-axis. The return of the security on the following day (day after statistic is taken) is on the y-axis.
Obviously one would go long the security when the statistic is <1 and short it when it is >1.

My question is - does this kind of strategy have a name? Has anyone else used data like this to decipher patterns?

Thanks.
 

Attachments

It's a "buy-at-open" (or "buy-at-prior-day's-close") "-and-hold-until-close" indicator-based strategy....?

Or it's a swing strategy, holding muliple days until you have a reverse signal ...?
 
Quote from abattia:

It's a "buy-at-open" (or "buy-at-prior-day's-close") "-and-hold-until-close" indicator-based strategy....?

Or it's a swing strategy, holding muliple days until you have a reverse signal ...?
More...

Its basically the second. Buy at end of day and hold until reverse signal obtained. Could be a day or a month. These are called "swing" strategies?
 
Quote from pistolpt:

I've enclosed a chart which holds one of the several keys to my investment strategy. This is based off of about 2,000 days of observations.
The prior day's value of a certain statistic is on the x-axis. The return of the security on the following day (day after statistic is taken) is on the y-axis.
Obviously one would go long the security when the statistic is <1 and short it when it is >1.

My question is - does this kind of strategy have a name? Has anyone else used data like this to decipher patterns?

Thanks.
More...

This appears to be an overbought oversold oscillator, based on mean reversion.

The .03 ranges should be broken out from thirtythirds into deciles.

Can you share a tradeslist? The effect of >1> is unclear, and there's so little data for each reading.
 
Since every bin along the horizontal axis has basically the same number of occurrences, it is highly unlikely the horizontal axis represents an oscillating indicator.

More likely it is measuring something using a fixed time period and/or tick count, very difficult to generate that kind of instance-distribution with any of the "normal" methods. Could also be random with a uniform distribution.
 
You must log in or register to reply here.
Back
Top