Module arima::estimate

Functions

• autofit
  TODO clean up Auto-fit an ARIMA model, guessing AR and MA orders. See fit for more details.
• fit
  Fit an ARIMA model. Returns the fitted coefficients. This method uses the L-BFGS algorithm and the conditional sum of squares (CSS) as the objective function.
• residuals
  Calculate residuals given a time series, an intercept, and ARMA parameters phi and theta. Any differencing and centering should be done before. Squaring and summing the residuals yields the conditional sum of squares (CSS), which can be used as an objective function to estimate the AR and MA parameters. The variance can be then estimated via CSS/(x.len()-phi.len()).