Abbreviation: LinTimeReg StndErr
Category: Regression
Input Parameters:
| Name | Range | Default |
| Time Series | Close | |
| Regression Periods | Int >= 3 | 10 |
Calculation:
sqrt ( [n*sum(Y*Y) – sum(Y)*sum(Y) – [ [n*sum(X*Y) – sum(X)*sum(Y)] ^ 2 ] / [n*sum(X*X) – sum(X)*sum(X) ] ] / [n*(n-1)] )
where
sum(X*X) = sum of X*X over the last n periods
sum(Y*Y) = sum of Y*Y over the last n periods
sum(X*Y) = sum of X*Y over the last n periods
sum(X) = sum of X over the last n periods
sum(Y) = sum of Y over the last n periods
X = Period Number
Y = Time Series
n = Regression Periods
Discussion:
Computes the standard error of estimate for the linear regression line. The standard error of estimate is simply the standard deviation of the vertical distance between the points and the linear regression line. Note that the linear regression line is the straight line that provides the best fit to the time series points (y axis) and their corresponding time period (x axis) over the last n periods.