Stata 15 help for mf_arfimaacf


[M-5] arfimaacf() -- Autocovariance functions


real colvector arfimaacf(real scalar n, real colvector phi, real colvector theta, real scalar d, real scalar v)


arfimaacf(n, phi, theta, d, v) computes the autocovariance function (ACF) of an autoregressive fractionally integrated moving-average (ARFIMA) process defined by the autoregressive parameters, phi, the moving-average parameters, theta, the fractional integration parameter, d, and the idiosyncratic error variance, v.


arfimaacf() returns a vector of length n+1, where the first element is the variance of the ARFIMA or ARMA process and elements k = 2, ..., n+1 are the autocovariances of the time-series process k-1 time units apart. The ACF of an ARMA process is obtained when d = 0.


arfimaacf(n, phi, theta, d, v): n: 1 x 1 phi: p x 1 theta: q x 1 d: 1 x 1 v: 1 x 1 result: n+1 x 1


The AR(p) and MA(q) polynomials defined by phi and theta must have roots outside the unit circle, and the two polynomials may not have common roots. The fractional integration parameter must be in (-1,1/2). The variance parameter, v, must be greater than zero.

Source code


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index