Significance of AIC differences for precipitation intensity distributions

Chain dependent models for daily precipitation typically model the occurrence process as a Markov chain and the precipitation intensity process using one of several probability distributions. It has been argued that the mixed exponential distribution is a superior model for the rainfall intensity process, since the value of its information criterion (Akaike information criterion or Bayesian information criterion) when fit to precipitation data is usually less than the more commonly used gamma distribution. The differences between the criterion values of the best and lesser models are generally small relative to the magnitude of the criterion value, which raises the question of whether these differences are statistically significant. Using a likelihood ratio statistic and nesting the gamma and mixed exponential distributions in a parent distribution, we show indirectly that generally the superiority of the mixed exponential distribution over the gamma distribution for modeling precipitation intensity is statistically significant. Comparisons are also made with a common-a gamma model, which are less informative.

[1]  Clifford M. Hurvich,et al.  Regression and time series model selection in small samples , 1989 .

[2]  David R. Cox,et al.  Further Results on Tests of Separate Families of Hypotheses , 1962 .

[3]  Environmental studies: Mathematical, computational, and statistical analysis , 1996 .

[4]  Dennis P. Lettenmaier,et al.  A Markov Renewal Model for Rainfall Occurrences , 1987 .

[5]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[6]  S. S. Wilks.  The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses , 1938 .

[7]  M. Parlange,et al.  Effects of an index of atmospheric circulation on stochastic properties of precipitation , 1993 .

[8]  John T. Kent,et al.  The underlying structure of nonnested hypothesis tests , 1986 .

[9]  Clive W. J. Granger,et al.  Comments on testing economic theories and the use of model selection criteria , 1995 .

[10]  P. Guttorp,et al.  A non‐homogeneous hidden Markov model for precipitation occurrence , 1999 .

[11]  M. Bhaskara Rao,et al.  Model Selection and Inference , 2000, Technometrics.

[12]  David R. Anderson,et al.  Null Hypothesis Testing: Problems, Prevalence, and an Alternative , 2000 .

[13]  W. Loh,et al.  A New Method for Testing Separate Families of Hypotheses , 1985 .

[14]  J. Hansen,et al.  Correcting low-frequency variability bias in stochastic weather generators , 2001 .

[15]  D. Wilks.  Interannual variability and extreme-value characteristics of several stochastic daily precipitation models , 1999 .

[16]  D. Rubin,et al.  Estimation and Hypothesis Testing in Finite Mixture Models , 1985 .

[17]  A. F. Smith,et al.  Statistical analysis of finite mixture distributions , 1986 .

[18]  David A. Woolhiser,et al.  A Stochastic Model of n-Day Precipitation , 1975 .

[19]  James W. Jones,et al.  WeatherMan: a utility for managing and generating daily weather data , 1994 .

[20]  R. Katz.  On Some Criteria for Estimating the Order of a Markov Chain , 1981 .

[21]  Richard W. Katz,et al.  Precipitation as a Chain-Dependent Process , 1977 .

[22]  Hirotugu Akaike,et al.  Analysis of cross classified data by AIC , 1978 .

[23]  N. Sugiura.  Further analysts of the data by akaike' s information criterion and the finite corrections , 1978 .

[24]  L. Pace,et al.  Best Conditional Tests for Separate Families of Hypotheses , 1990 .

[25]  P. Thornton,et al.  A rainfall generator for agricultural applications in the tropics , 1993 .

[26]  H. Bozdogan,et al.  Akaike's Information Criterion and Recent Developments in Information Complexity. , 2000, Journal of mathematical psychology.

[27]  T. A. Buishand,et al.  Some remarks on the use of daily rainfall models , 1978 .

[28]  F. Moreno,et al.  Regionalization of daily precipitation stochastic model parameters. Application to the guadalquivir valley in Southern Spain , 1999 .

[29]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[30]  N. Ison,et al.  Wet Period Precipitation and the Gamma Distribution. , 1971 .

[31]  Peter Guttorp,et al.  Stochastic modeling of rainfall , 1996 .

[32]  M. Parlange,et al.  Overdispersion phenomenon in stochastic modeling of precipitation , 1998 .

[33]  James P. Hughes,et al.  A class of stochastic models for relating synoptic atmospheric patterns to regional hydrologic phenomena , 1994 .

[34]  M. Parlange,et al.  An Extended Version of the Richardson Model for Simulating Daily Weather Variables , 2000 .