scholarly journals Mixed Portmanteau Test for Diagnostic Checking of Time Series Models

2014 ◽  
Vol 2014 ◽  
pp. 1-8 ◽  
Author(s):  
Sohail Chand ◽  
Shahid Kamal
Keyword(s):  
Time Series ◽  
Monte Carlo ◽  
Goodness Of Fit ◽  
Model Building ◽  
Main Idea ◽  
Dependence Structure ◽  
Time Series Models ◽  
Goodness Of Fit Tests ◽  
Diagnostic Checking ◽  
Portmanteau Tests

Model criticism is an important stage of model building and thus goodness of fit tests provides a set of tools for diagnostic checking of the fitted model. Several tests are suggested in literature for diagnostic checking. These tests use autocorrelation or partial autocorrelation in the residuals to criticize the adequacy of fitted model. The main idea underlying these portmanteau tests is to identify if there is any dependence structure which is yet unexplained by the fitted model. In this paper, we suggest mixed portmanteau tests based on autocorrelation and partial autocorrelation functions of the residuals. We derived the asymptotic distribution of the mixture test and studied its size and power using Monte Carlo simulations.

scholarly journals Dependent SiZer: Goodness-of-Fit Tests for Time Series Models

2004 ◽  
Vol 31 (8) ◽  
pp. 999-1017 ◽  
Author(s):  
Cheolwoo Park ◽  
J. S. Marron ◽  
Vitaliana Rondonotti
Keyword(s):  
Time Series ◽  
Goodness Of Fit ◽  
Time Series Models ◽  
Goodness Of Fit Tests

scholarly journals GOODNESS-OF-FIT TESTS FOR MULTIVARIATE COPULA-BASED TIME SERIES MODELS

2016 ◽  
Vol 33 (2) ◽  
pp. 292-330 ◽  
Author(s):  
Betina Berghaus ◽  
Axel Bücher
Keyword(s):  
Time Series ◽  
Goodness Of Fit ◽  
Mixing Time ◽  
Parametric Bootstrap ◽  
Stationary Time Series ◽  
Time Series Models ◽  
Finite Sample ◽  
Copula Functions ◽  
Cross Sectional ◽  
Goodness Of Fit Tests

In recent years, stationary time series models based on copula functions became increasingly popular in econometrics to model nonlinear temporal and cross-sectional dependencies. Within these models, we consider the problem of testing the goodness-of-fit of the parametric form of the underlying copula. Our approach is based on a dependent multiplier bootstrap and it can be applied to any stationary, strongly mixing time series. The method extends recent i.i.d. results by Kojadinovic et al. (2011) and shares the same computational benefits compared to methods based on a parametric bootstrap. The finite-sample performance of our approach is investigated by Monte Carlo experiments for the case of copula-based Markovian time series models.

scholarly journals A Monte Carlo Simulation Study on the Power of Autocorrelation Tests for ARMA Models

2019 ◽  
Vol 16 (3) ◽  
pp. 59-67
Author(s):  
Zachary Wenning ◽  
Emily Valenci
Keyword(s):  
Time Series ◽  
Monte Carlo ◽  
Goodness Of Fit ◽  
Arma Model ◽  
Small Sample ◽  
Arma Models ◽  
Real World Data ◽  
Data Set ◽  
Monte Carlo Simulation Study ◽  
Portmanteau Tests

It is often the case when assessing the goodness of fit for an ARMA time series model that a portmanteau test of the residuals is conducted to assess residual serial correlation of the fitted ARMA model. Of the many portmanteau tests available for this purpose, one of the most famous and widely used is a variant of the original Box-Pierce test, the Ljung-Box test. Despite the popularity of this test, however, there are several other more modern portmanteau tests available to assess residual serial autocorrelation of the fitted ARMA model. These include two portmanteau tests proposed by Monti and Peña and Rodríguez. This paper focuses on the results of a power analysis comparing these three different portmanteau tests against different fits of ARMA - derived time series, as well as the behavior of the three different test statistics examined when applied to a real-world data set. We confirm that for situations in which the moving average component of a fitted ARMA model is underestimated or when the sample size is small, the portmanteau test proposed by Monti is a viable alternative to the Ljung-Box test. We show new evidence that the Peña and Rodríguez may also be a viable option for testing for residual autocorrelation for data with small sample sizes. KEYWORDS: Time Series; Monte Carlo; ARMA Models; Power; Simulation; Autocorrelation Tests; Portmanteau Tests; Monti; Ljung-Box; Peña and Rodríguez

Sign in / Sign up

Export Citation Format

Share Document