A full ARMA model for counts with bounded support and its application to rainy-days time series

Motivated by a large dataset containing time series of weekly number of rainy days collected over two thousand locations across Europe and Russia for the period 2000–2010, we propose a new class of ARMA-like model for time series of bounded counts, which can also handle extra-binomial variation. We abbreviate this model as bvARMA, as it is based upon a novel operation referred to as binomial variation. After having discussed important stochastic properties and proposed a model-fitting approach relying on maximum likelihood estimation, we apply the bvARMA model family to the rainy-days time series. Results show that both bvAR and bvMA models are adequate and exhibit a similar performance. Furthermore, bvARMA results outperform those obtained by fitting ordinary discrete ARMA (NDARMA) models of the same order.

[1]  Christian H. Weií.  Serial dependence of NDARMA processes , 2013 .

[2]  Pedro M. Valero-Mora,et al.  ggplot2: Elegant Graphics for Data Analysis , 2010 .

[3]  T. J. Chang,et al.  Modeling of Sequences of Wet and Dry Days by Binary Discrete Autoregressive Moving Average Processes , 1984 .

[4]  T. Asefa,et al.  Improving Short-Term Urban Water Demand Forecasts with Reforecast Analog Ensembles , 2016 .

[5]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[6]  Hadley Wickham,et al.  ggplot2 - Elegant Graphics for Data Analysis (2nd Edition) , 2017 .

[7]  Rainer Göb,et al.  Measuring serial dependence in categorical time series , 2008 .

[8]  Robert Lund,et al.  A new look at time series of counts , 2009 .

[9]  T. J. Chang,et al.  Simulation Models of Sequences of Dry and Wet Days , 1989 .

[10]  Rob J Hyndman,et al.  Theory & Methods: Non‐Gaussian Conditional Linear AR(1) Models , 2000 .

[11]  Christian H. Weiß.  Serial dependence of NDARMA processes , 2013, Comput. Stat. Data Anal..

[12]  N. Ehelepola,et al.  A study of the correlation between dengue and weather in Kandy City, Sri Lanka (2003 -2012) and lessons learned , 2015, Infectious Diseases of Poverty.

[13]  Wooi Chen Khoo,et al.  Modeling time series of counts with a new class of INAR(1) model , 2017 .

[14]  Antonio Salmerón,et al.  Continuous Bayesian networks for probabilistic environmental risk mapping , 2016, Stochastic Environmental Research and Risk Assessment.

[15]  Anders Moberg,et al.  Daily dataset of 20th‐century surface air temperature and precipitation series for the European Climate Assessment , 2002 .

[16]  Time series formed from the superposition of discrete renewal processes , 1989 .

[17]  D. Karlis,et al.  INAR(1) modeling of overdispersed count series with an environmental application , 2008 .

[18]  Fw Fred Steutel,et al.  Discrete analogues of self-decomposability and stability , 1979 .

[19]  B. Pohl,et al.  Fewer rainy days and more extreme rainfall by the end of the century in Southern Africa , 2017, Scientific Reports.

[20]  Guillaume Lacombe,et al.  Uncovering consistencies in Indian rainfall trends observed over the last half century , 2014, Climatic Change.

[21]  A. Berchtold.  High-order extensions of the Double Chain Markov Model , 2002 .

[22]  G. G. S. Pegram,et al.  An autoregressive model for multilag Markov chains , 1980, Journal of Applied Probability.

[23]  Peter A. W. Lewis,et al.  STATIONARY DISCRETE AUTOREGRESSIVE‐MOVING AVERAGE TIME SERIES GENERATED BY MIXTURES , 1983 .

[24]  Maria Eduarda Silva,et al.  Bivariate binomial autoregressive models , 2014, J. Multivar. Anal..

[25]  W. Zucchini,et al.  Hidden Markov Models for Time Series: An Introduction Using R , 2009 .

[26]  Peter Thyregod,et al.  Integer Valued Autoregressive Models for Tipping Bucket Rainfall Measurements , 1999 .