Incorporate the ARtCensReg function to fit a
univariate censored linear regression model with autoregressive errors
considering Student-t innovations.
Generic functions predict, print,
summary, and plot have methods for objects
given as an output of ARCensReg and ARtCensReg
functions.
Function residuals was incorporated to compute the
conditional and quantile residuals for objects inheriting from class
ARpCRM or ARtpCRM, given as an output of ARCensReg and
ARtCensReg function, respectively.
Function plot is also valid for objects returned by
residuals. This procedure returns the following plots for
the quantile residuals: residual vs. time, an autocorrelation plot, a
histogram, and a Quantile-Quantile (Q-Q) plot.
Argument pit was substituted by phi in
the function rARCens. Please see the documentation for more
information.
Function rARCens was modified to simulate datasets
with Student-t innovations.
Some modifications in the arguments of the InfDiag
function were made. The generic function plot is available
for outputs of function InfDiag.
Incorporate the rARCens function to simulate
censored autoregressive data.
Incorporate the InfDiag function to perform
influence diagnostic by a local influence approach with three possible
perturbation schemes: response perturbation, scale matrix perturbation,
or explanatory variable perturbation.
The Phosphorus concentration data was added.