vcov.arfima {arfima} | R Documentation |
Extracts the variance-covariance matrices (one or two for each mode) from a
fitted arfima
object.
## S3 method for class 'arfima' vcov(object, type = c("b", "o", "e"), cor = FALSE, digits = max(4, getOption("digits") - 3), tapprox = FALSE, summ = FALSE, ...)
object |
A fitted |
type |
Which type of covariance matrix to return: "o" is the observed matrix (from solving the Hessian), "e" is the expected matrix (from solving the information matrix), and "b" is both. |
cor |
Whether or not the correlation matrix should be returned instead. |
digits |
The number of digits to print. |
tapprox |
Whether or not to use an approximation to find the expected matrix. Highly recommended to be FALSE, as it takes much longer, and is an approximation. |
summ |
Whether the call is from the |
... |
Optional arguments, currently not used. |
A list of lists (one for each mode) with components observed
and/or expected
.
JQ (Justin) Veenstra
Veenstra, J.Q. Persistence and Antipersistence: Theory and Software (PhD Thesis)
set.seed(1234) sim <- arfima.sim(1000, model = list(dfrac = 0.4, phi = .8, theta = -0.5)) fit1 <- arfima(sim, order = c(1, 0, 1), back=TRUE) fit2 <- arfima(sim, order = c(1, 0, 1), lmodel = "g", back=TRUE) fit3 <- arfima(sim, order = c(1, 0, 1), lmodel = "h", back=TRUE) fit1 fit2 fit3 vcov(fit1) vcov(fit2) vcov(fit2)