Fits an autoregressive model of order p for the horizon-h
target and returns the OLS coefficients and residuals.
Arguments
- y
Numeric vector of the target variable.
- h
Positive integer; forecast horizon.
- p
Non-negative integer; AR lag order.
Value
A list with components:
- a_hat
Coefficient vector (intercept first).
- res
Residual vector.
Examples
y <- arima.sim(list(ar = 0.7), n = 200)
ar_fit <- estimate_ar_res(y, h = 1, p = 1)
ar_fit$a_hat
#> [,1]
#> [1,] 0.1027578
#> [2,] 0.7011676
