lin1_CI: Confidence regions for linar AR(1) processes with intercept...

In ChrisKust/rexpar: Simplicial Depth for Explosive Autoregressive Processes

Description

This function generates confidence intervals for linear AR(1) processes without intercept and explosion based on simplicial depth for AR processes.

Usage

lin1_CI(y, level, plots = FALSE, notion = "dS", eps = NULL)

Arguments

y	A observed series from an linear AR(1) process with intercept.
level	A value in (0,1) defining the level of the confidence regions to evaluate.
plots	A swich to turn on and off plots of the resulting region. Use plots = TRUE to turn on plots and plots = FALSE (default) to suppress plotting.
notion	Here the applied test statistic for the calculation of confidence intervals is specified. The available notions are "dS" = full simplicial depth "dS1" = reduced notion without overlapping of residuals "dS2" = reduced notion based on large distance residuals "dS3" = reduced notion with overlapping residuals
eps	This allows to define a shifting of the candidate paramters by a constant eps to avoid evaluations on candidates defined by roots of the residuals.

Details

The theoretical background can be found in Kustosz, Mueller and Wendler (2016) and Kustosz and Mueller (2014). Details on the implementation can be found in Kustosz (2016).

Value

par	A vector evaluated points to calculate the confidence region by evaluation of the depth tests.
inCI	A binary vector indicating, if a parameter in par is in the confidence interval or not. Thereby the test decision is reported. Hence, inCI = 0 donotes a parameter being inside of the confidence set and inCI = 1 denotes a parameter which is not.

Author(s)

Kustosz, Christoph

References

Kustosz, C. (2016). Depth based estimators and tests for autoregressive processes with application. Ph. D. thesis. TU Dortmund.

Kustosz C., Mueller Ch. H. and Wendler M. (2016). Simplified Simplicial Depth for Regression and Autoregressive Growth Processes. Journal of Statistical Planning and Inference. In press.

Kustosz C. and Mueller Ch. H. (2014). Analysis of crack growth with robust distribution- free estimators and tests for nonstationary autoregressive processes. Statistical Papers 55, 125-140.

See Also

dS1_lin1_test,dS2_lin1_test,dS3_lin1_test,dS_lin1_test,lin1_theta

Examples

y <- RandomARMod_lin2(100, 0, 1.001, 15, "0")
lin1_CI(y, 0.05) -> CI
ci_result <- c(min(CI$par[!CI$inCI]),max(CI$par[!CI$inCI]))
print(ci_result)

ChrisKust/rexpar documentation built on May 6, 2019, 11:48 a.m.