AccScience Publishing / AJWEP / Volume 18 / Issue 1 / DOI: 10.3233/AJW210005
RESEARCH ARTICLE

Modelling Unusual Behaviour of Rainfall Using  Truncated GEV Distribution in a Mixture Framework

R.S. Jagtap1* U.V. Naik-Nimbalkar2
Show Less
1 Central Water and Power Research Station, Khadakwasla, Pune – 411024, India
2 Department of Statistics & Centre for Advanced Studies in Statistics, Department of Statistics & Centre for Advanced Studies in Statistics
AJWEP 2021, 18(1), 31–41; https://doi.org/10.3233/AJW210005
Submitted: 1 September 2018 | Revised: 15 December 2020 | Accepted: 15 December 2020 | Published: 25 January 2021
© 2021 by the Author(s). This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution -Noncommercial 4.0 International License (CC-by the license) ( https://creativecommons.org/licenses/by-nc/4.0/ )
Abstract

A truncated generalised extreme value (GEV) distribution in a mixture framework is proposed for the  analysis of abnormal rare events in heterogeneous data representing environmental phenomena. The proposed  extremal mixture model produced a better understanding of the extremal rainfall behaviour in the Mula-MuthaBhima subbasin in India. It gave some realistic extrapolation of quantiles corresponding to a very low probability  of exceedance useful in water resources planning and design of civil infrastructure. The proposed model could  be useful for the class of problems characterising extreme events and heterogeneity in fields like hydrology,  environment and so on.

Keywords
Extreme values
heterogeneity
maximum likelihood estimate
mixture modelling
quantile
truncated GEV distribution.
Conflict of interest
The authors declare they have no competing interests.
References

Akaike, H. (1974). A new look at the statistical model  identification. IEEE Transactions on Automatic Control,  19: 716-723.

Azzalini, A. (1985). A class of distributions which includes  the normal ones. Scandinavian Journal of Statistics, 12: 171-178.

Behrens, C.N., Lopes, H.F. and D. Gamerman (2004).  Bayesian analysis of extreme events with threshold  estimation. Statistical Modeling, 4(3): 227-244.

Cannon, A.J. (2010). A flexible nonlinear modelling  framework for non-stationary generalized extreme value  analysis in hydroclimatology. Hydrological Processes,  24: 673-685.

Coles, S. (2001). An Introduction to Statistical Modelling of  Extreme Values. Springer: London. 

Escalante-Sandoval, C. (2007). A Mixed distribution with  EV1 and GEV components for analyzing heterogeneous  samples. Engineering, Research and Technology, VIII.  3: 123-133 pp.

Hosking, J.R.M. and J.R. Wallis (1997). Regional Frequency  Analysis: An Approach Based on L-Moments. Cambridge  University Press, Cambridge, UK, 224 pp.

Ma, Y. and M.G. Genton (2004). Flexible class of skewsymmetric distribution. Scandinavian Journal of Statistics,  31, Issue 3, Sept.

McLachlan, G.J. and D. Peel (2000). Finite Mixture Models.  Wiley, New York.

Mendes, B.V.M. and H.F. Lopes (2004). Data driven estimates  for mixtures. In: Computational Statistics & Data Analysis,  47: 583-598. Elsevier.

Pickands, J. (1975). Statistical inference using extreme order  statistics. Annals of Statistics, 3: 119-131.

R Core Team (2015). R: A language and environment  for statistical computing. R Foundation for Statistical  Computing, Vienna, Austria, URL: https://www.R-project. org/.

Reeve, D.E. (1996). Estimation of extreme Indian monsoon  rainfall. International Journal of Climatology, 16: 105- 112, Shorter contribution.

Roxy, M.K., Ghosh, S., Amey Pathak, A., Athulya, R.,  Mujumdar, M., Murtugudde,R., Terray P. and M. Rajeevan(2017). A threefold rise in widespread extreme rain events  over central India. Nature Communications, 8: 708, DOI:  10.1038/s41467-017-00744-9.

Smith, R.L. (1985). Maximum likelihood estimation in a class  of non-regular cases. Biometrika, 72: 67–90.

Share
Back to top
Asian Journal of Water, Environment and Pollution, Electronic ISSN: 1875-8568 Print ISSN: 0972-9860, Published by AccScience Publishing