Math wizards, is ATR superior to Beta?

D

Daal

Lets say you take a stock ATR divide by the stock price and multiply by 100(To get percentage terms), you do the same for the SPY and compare the ratio of the stock ATR Vol to SPY Vol, you will get a number that in theory is similar in nature to beta(The actual number differ a lot)

Which one would be superior in forecasting that stock price if the SPY for example, drops by 10%?
 
Quote from Daal:

Lets say you take a stock ATR divide by the stock price and multiply by 100(To get percentage terms), you do the same for the SPY and compare the ratio of the stock ATR Vol to SPY Vol, you will get a number that in theory is similar in nature to beta(The actual number differ a lot)

Which one would be superior in forecasting that stock price if the SPY for example, drops by 10%?
More...
================
Sector trends;
& 200 dma probably would help much more than either.

One stock is a risky thing anyway:D
 
In THEORY, beta gives you more information, because it also takes correlation into account. An ATR ratio just looks at relative volatilities. So, a stock with beta of 0 could have the same ATR.

HOWEVER, beta doesn't necessarily predict movement. That same zero beta stock may still bounce around just as much as the index, but the zero beta just tells you that it won't necessarily go in the same direction.

Of course, smarter people than me have shown that beta is a far from perfect measure, as it changes over time. Therefore, a stock with 0.5 beta based on past data may change to 0.8 beta in the future.
 
Quote from drm7:

In THEORY, beta gives you more information, because it also takes correlation into account. An ATR ratio just looks at relative volatilities. So, a stock with beta of 0 could have the same ATR.

HOWEVER, beta doesn't necessarily predict movement. That same zero beta stock may still bounce around just as much as the index, but the zero beta just tells you that it won't necessarily go in the same direction.

Of course, smarter people than me have shown that beta is a far from perfect measure, as it changes over time. Therefore, a stock with 0.5 beta based on past data may change to 0.8 beta in the future.
More...

Yeah, this seems right. The difference might be that in period of relative tranquility the beta model will do better but in a crisis(where correlations go to 1) the ATR model would do better

I supposed for trading in short period(daytrading) beta is superior because its unlikely to change much in that timeframe. Over long period(position trading, investing), relying on beta gets riskier
 
You must log in or register to reply here.
Back
Top