Hi @David Harper CFA FRM
I am a bit confused by the description that partial autocorrelation function as the coefficient if the most distant term after regressing the current in the past series. Don't we have only one series here, and therefore, only 1 sample point for each independent variable?
For example, say series (X3 at time -3 = -1.2, X2 at time -2 = -0.4, X1 at time -1 = 1, and the current value of the series at time 0 = 0.5). Then we would have:
X3 = -1.2
X2 = -0.4
X1 = +1.0
Y = 0.5
Then how would we regress Y on the Xs here? Can you provide a small example in excel illustrating the determination of the partial autocorrelation function?
I am a bit confused by the description that partial autocorrelation function as the coefficient if the most distant term after regressing the current in the past series. Don't we have only one series here, and therefore, only 1 sample point for each independent variable?
For example, say series (X3 at time -3 = -1.2, X2 at time -2 = -0.4, X1 at time -1 = 1, and the current value of the series at time 0 = 0.5). Then we would have:
X3 = -1.2
X2 = -0.4
X1 = +1.0
Y = 0.5
Then how would we regress Y on the Xs here? Can you provide a small example in excel illustrating the determination of the partial autocorrelation function?