Currently displaying 1 – 20 of 44

Showing per page

Order by Relevance | Title | Year of publication

An estimator for parameters of a nonlinear nonnegative multidimensional AR(1) process

Jiří Anděl — 1998

Applications of Mathematics

Let 𝕖 t = ( e t 1 , , e t p ) ' be a p -dimensional nonnegative strict white noise with finite second moments. Let h i j ( x ) be nondecreasing functions from [ 0 , ) onto [ 0 , ) such that h i j ( x ) x for i , j = 1 , , p . Let 𝕌 = ( u i j ) be a p × p matrix with nonnegative elements having all its roots inside the unit circle. Define a process 𝕏 t = ( X t 1 , , X t p ) ' by X t j = u j 1 h 1 j ( X t - 1 , 1 ) + + u j p h p j ( X t - 1 , p ) + e t j for j = 1 , , p . A method for estimating 𝕌 from a realization 𝕏 1 , , 𝕏 n is proposed. It is proved that the estimators are strongly consistent.

Statistical analysis of periodic autoregression

Jiří Anděl — 1983

Aplikace matematiky

Methods for estimating parameters and testing hypotheses in a periodic autoregression are investigated in the paper. The parameters of the model are supposed to be random variables with a vague prior density. The innovation process can have either constant or periodically changing variances. Theoretical results are demonstrated on two simulated series and on two sets of real data.

Periodic autoregression with exogenous variables and periodic variances

Jiří Anděl — 1989

Aplikace matematiky

The periodic autoregressive process with non-vanishing mean and with exogenous variables is investigated in the paper. It is assumed that the model has also periodic variances. The statistical analysis is based on the Bayes approach with a vague prior density. Estimators of the parameters and asymptotic tests of hypotheses are derived.

On multiple periodic autoregression

Jiří Anděl — 1987

Aplikace matematiky

The model of periodic autoregression is generalized to the multivariate case. The autoregressive matrices are periodic functions of time. The mean value of the process can be a non-vanishing periodic sequence of vectors. Estimators of parameters and tests of statistical hypotheses are based on the Bayes approach. Two main versions of the model are investigated, one with constant variance matrices and the other with periodic variance matrices of the innovation process.

On non-nested regression models

Jiří Anděl — 1993

Commentationes Mathematicae Universitatis Carolinae

A generalization of a test for non-nested models in linear regression is derived for the case when there are several regression models with more regressors.

Page 1 Next

Download Results (CSV)