No. of Recommendations: 2
I have a quick question for Elan and Jim.
As per the old Motley Fool posts from 1999, DH10 RRS 126-2s screen definition said: The regression line slope over the last 126 trading days, using log-price versus trading day.
And in one of the posts from Jim a year ago he mentioned the methodology as:
Take the prices for the last 6 months for the first stock, let's say in column A.
Take the log of all those prices = LN (a1) in column B.
Calculate the slope, =SLOPE (B1:B126,A1:A126).
As per Jim’s explanation above, the regression line slope is using log-price versus the “actual price” instead of trading days.
Which is the correct way to calculate the regression slope for DH10 RRS 126-2s?
Thanks in advance.
AJ
No. of Recommendations: 2
Which is the correct way to calculate the regression slope
I would have to think it's log-price vs trading day. I suspect Jim left out a column as he was thinking through the calculation. I suspect 'A' would be dates, 'B' = price, 'C' = LN(B). Then m = SLOPE(A, C).
But I'd certainly like to know if I'm wrong about that.
John
No. of Recommendations: 3
I would have to think it's log-price vs trading day. I suspect Jim left out a column as he was thinking through the calculation. I suspect 'A' would be dates, 'B' = price, 'C' = LN(B). Then m = SLOPE(A, C).
But I'd certainly like to know if I'm wrong about that.
I believe you're absolutely right.
Elan
