partialAR (version 1.0.3)

likelihood_ratio.par: Computes log likelihood ratio for partial autoregressive model

Description

Computes the log likelihood ratio for the partially autoregressive model.

First, a fit is performed for the specified null model. Then, a fit is performed for the alternative model that the sequence is partially autoregressive. The likelihood scores are computed for both models, and the log likelihood ratio is returned.

Usage

likelihood_ratio.par(X, robust = FALSE, null_model = c("rw", "ar1"), 
  opt_method = c("css", "fkf", "ss"), nu = par.nu.default())

Arguments

X

The numeric vector or zoo vector to which the partially autoregressive model is being fit.

robust

If TRUE, then errors are assumed to follow a t-distribution with nu degrees of freedom. If FALSE, then errors are assumed to follow a normal distribution. Default: FALSE

null_model

Specifies the null hypothesis:

• "rw"

{Pure random walk (e.g., sigma_M = 0)} "ar1" {Pure autoregressive (e.g., sigma_R = 0)}

Value

• A numeric value representing the log likelihood ratio

item

• opt_method
• nu

itemize

• "ss"Steady-state Kalman filter with normally distributed errors
• "css"Steady-state Kalman filter with normally distributed errors, coded in C++
• "fkf"Traditional Kalman filter of thefkfpackage

code

TRUE

References

Clegg, Matthew. Modeling Time Series with Both Permanent and Transient Components using the Partially Autoregressive Model. Available at SSRN: http://ssrn.com/abstract=2556957

See Also

fit.par