Influence diagnostics in log-linear integer-valued GARCH models

Integer-valued generalized autoregressive conditional heteroscedasticity (GARCH) models have played an important role in time series analysis of count data. To model negatively autocorrelated time series and to accommodate covariates without restrictions, the log-linear integer-valued GARCH model has recently been proposed as an alternative to the existing models. In this paper, we study a local influence diagnostic analysis in the log-linear integer-valued GARCH models. The slope-based diagnostic and stepwise curvature-based diagnostics in a framework of the modified likelihood displacement are proposed. Under five perturbation schemes the corresponding local influence measures are derived. Two simulated data sets and a real-world example are analyzed to illustrate our method. In addition, the fitted model for this example has a negative coefficient for one of the two covariates, which is particularly illustrative of the extra flexibility of the considered model.

[1]  A. J. Lawrance,et al.  Deletion Influence and Masking in Regression , 1995 .

[2]  Sanford Weisberg,et al.  Directions in Robust Statistics and Diagnostics , 1991 .

[3]  Fukang Zhu,et al.  Estimation and testing for a Poisson autoregressive model , 2011 .

[4]  Fei Chen,et al.  Outlier Detection in Time Series Models Using Local Influence Method , 2012 .

[5]  Fukang Zhu,et al.  Modeling time series of counts with COM-Poisson INGARCH models , 2012, Math. Comput. Model..

[6]  Stavros Degiannakis,et al.  ARCH Models for Financial Applications , 2010 .

[7]  L. K. Hotta,et al.  Influential observations in GARCH models , 2012 .

[8]  K. Fokianos.  Count Time Series Models , 2012 .

[9]  Lei Shi,et al.  Local influence in principal components analysis , 1997 .

[10]  R. Cook.  Assessment of Local Influence , 1986 .

[11]  Viliam Makis,et al.  Highway accident modeling and forecasting in winter , 2014 .

[12]  Fukang Zhu.  Modeling overdispersed or underdispersed count data with generalized Poisson integer-valued GARCH models , 2012 .

[13]  K. Fokianos,et al.  Interventions in log-linear Poisson autoregression , 2012 .

[14]  Dimitris Karlis,et al.  On composite likelihood estimation of a multivariate INAR(1) model , 2013 .

[15]  Dag Tjøstheim,et al.  Some recent theory for autoregressive count time series , 2012 .

[16]  Xibin Zhang,et al.  Assessment of Local Influence in GARCH Processes , 2004 .

[17]  J. Zakoian,et al.  GARCH Models: Structure, Statistical Inference and Financial Applications , 2010 .

[18]  Lei Shi,et al.  Stepwise local influence analysis , 2011, Comput. Stat. Data Anal..

[19]  Roland Fried,et al.  ROBUST FITTING OF INARCH MODELS , 2014 .

[20]  Fukang Zhu.  A negative binomial integer‐valued GARCH model , 2010 .

[21]  Yat Sun Poon,et al.  Conformal normal curvature and assessment of local influence , 1999 .

[22]  M. King,et al.  Influence Diagnostics in Generalized Autoregressive Conditional Heteroscedasticity Processes , 2005 .

[23]  Calyampudi Radhakrishna Rao,et al.  Time series analysis : methods and applications , 2012 .

[24]  Shuangzhe Liu,et al.  On diagnostics in conditionally heteroskedastic time series models under elliptical distributions , 2004, Journal of Applied Probability.

[25]  Tim Dunne,et al.  Diagnostics for Regression-Arma Time Series , 1991 .

[26]  Fukang Zhu.  Zero-inflated Poisson and negative binomial integer-valued GARCH models , 2012 .

[27]  Sik-Yum Lee,et al.  Local influence for incomplete data models , 2001 .

[28]  Eric Moulines,et al.  Ergodicity of observation-driven time series models and consistency of the maximum likelihood estimator , 2012, 1210.4739.

[29]  Dominique Lord,et al.  Extension of Negative Binomial GARCH Model , 2012 .

[30]  Fukang Zhu,et al.  Diagnostic checking integer-valued ARCH(p) models using conditional residual autocorrelations , 2010, Comput. Stat. Data Anal..

[31]  Xibin Zhang,et al.  Influence diagnostics for multivariate GARCH processes , 2010 .

[32]  Xizhi Wu,et al.  Second‐Order Approach to Local Influence , 1993 .

[33]  Dehui Wang,et al.  A mixture integer-valued ARCH model , 2010 .

[34]  Alain Latour,et al.  Integer‐Valued GARCH Process , 2006 .

[35]  R. M. Loynes,et al.  Local influence: a new approach , 1993 .

[36]  Geert Wets,et al.  Studying the effect of weather conditions on daily crash counts using a discrete time-series model. , 2008, Accident; analysis and prevention.

[37]  Konstantinos Fokianos,et al.  Log-linear Poisson autoregression , 2011, J. Multivar. Anal..