Rainfall Prediction in Statistical Downscaling Using Tweedie Compound Response and Lasso Penalty

Authors

  • Marufah Hayati  Department of Statistic, The University of Nahdlatul Ulama Lampung, Sukadana, Indonesia
  • Reni Permatasari  Department of Statistic, The University of Nahdlatul Ulama Lampung, Sukadana, Indonesia

DOI:

https://doi.org/10.32628/IJSRSET23103121

Keywords:

Tweedie Compound, Poisson-Gamma, Statistical Downscaling, Lasso, Rainfall.

Abstract

Statistical Downscaling (SD) is a technique in climatology to analyze the relationship between large-scale (global) data and small-scale (local) data using statistical modeling. The SD technique is used to overcome the inability of global scale data in the form of the General Circulation Model (GCM) as a low resolution predictor to predict local scale climatic conditions in the form of high resolution rainfall as a direct response. Rainfall consists of two components, namely continuous and discrete. The continuous component describes the intensity of rainfall while the discrete component describes the occurrence of rain. both components have an important role in predicting rainfall so it is necessary to choose the right distribution. One distribution that is able to handle both rain components is the mixed Tweedie distribution, namely the Gamma and Poisson distribution, hereinafter referred to as the Tweedie compound. GCM generally has multicollinearity problems in SD modeling. This can be handled using the Lasso penalty. This study aims to predict rainfall and rainfall events by taking into account the multicollinearity problem in the model for locations on different plains. Based on the research results, it was found that Cigugur Station from the highland gets the smallest RMSEP value and the biggest r-correlation. This model is not good enough to use for moderate plains rainfall data.

References

  1. Sutikno. 2008. Statistical downscaling luaran GCM dan pemanfaatannya untuk peramalan produksi padi, Disertation, IPB University.
  2. M.M. Hasan, P.K. Dunn. A simple Poisson–gamma model for modelling rainfall occurrence and amount simultaneously, Agric. Forest Meteorol. 150. 1319–1330, 2010.
  3. N.C. Dzupire, P. Ngare, L. Odongo, A Poisson-Gamma model for zero inflated rainfall data, J. Probab. Stat. (2018), 1012647
  4. F. Novkaniza, M. Hayati, B. Sartono, K. Anwar Notodiputro, Fused lasso for modeling monthly raifall in Indramayu sub distric West Java Indonesia, IOP Conf. Ser.: Earth Environ. Sci. 187 (2018), 012046. 2018.
  5. A. Muslim, M. Hayati, B. Sartono, K.A. Notodiputro, A combined modeling of generalized linear mixed model and LASSO techniques for analizing monthly rainfall data, IOP Conf. Ser.: Earth Environ. Sci. 187 (2018), 012044
  6. M. Hayati and A. Muslim. Generalized Linear Mixed Model and Lasso Regularization for Statistical Downscaling. ENTHUSIASTIC: International Journal of Statistics and Data Science. vol. 1, No. 1, April 2021, pp. 36-52. 2021.
  7. WH Bonat, C.C. Kokonendji. Flexible Tweedie regression models for continuous data. J Stat Comput Simul. 87(11):2138–2152. doi:10.1080/00949655.2017.1318876, 2017.
  8. P.K. Dunn, Occurrence and quantity of precipitation can be modeled simultaneously. Int. J. Climatol. 24 (2004), 1231–1239, 2004.
  9. M Hayati, A. H. Wigena, A Djuraidah, A. Kurnia. A New Approach To Statistical Downscaling Using Tweedie Compound Poisson Gamma Response And Lasso Regularization. Commun. Math. Biol. Neurosci, 2021:60. ISSN: 2052-254, 2021.
  10. M Hayati, A. H. Wigena, A Djuraidah, A. Kurnia. The Spatio-Temporal Model For The Tweedie Compound Poisson Gamma Response In Statistical Downscaling. J. Math. Comput. Sci. 2022, 12:35, ISSN: 1927-5307, 2022,.
  11. E. Zorita, H. von Storch, The analog method as a simple statistical downscaling technique: comparison with more complicated methods. J. Climate, 12 , 2474-2489, 1999.
  12. W. Qian, Y. Yang, H. Zou, Tweedie’s compound poisson model with grouped elastic net, J. Comput. Graph. Stat. 25 , 606–625, 2016.
  13. Y. Yang, R. Luo, R Khorsidi, Y Liu, Inferring tweedie compound poisson mixed models with adversarial variational methods, Presented at NIPS Workshop on Advances in Approximate Bayesian Inference, 2017.
  14. Ma R. J?rgensen B. Nested Generalized Linier Mixed Models: an ortodhox Best Linear Unbiased Predictor Approach, J.R. Statist.Soc.B. 69( 4) , 625-641, 2006.
  15. Y.S. Trzaska, E. Schnarr. A review of downscaling methods for climate change projection,. 2014 https://www.climatelinks.org/resources/review-downscaling-methods-climate-change-projections
  16. P.K. Dunn,Tweedie: Tweedie exponential family models; R Package. Version 2.2.1; R Core Team: Vienna, Austria, 2014.
  17. P.K. Dunn, Occurrence and quantity of precipitation can be modeled simultaneously. Int. J. Climatol. 24, 1231–1239, 2004.
  18. P.K. Dunn, G.K. Smyth, Series evaluation of Tweedie exponential dispersion model densities, Stat. Comput. 15, 267–280, 2005.
  19. T. Hastie, R. Tibshirani, J. Friedman. The elements of statistical learning: data mining, inference, and prediction. 2009. https://web.stanford.edu/~hastie/ElemStatLearn

International Journal of Scientific Research in Science, Engineering and Technology

Downloads

Published

2023-06-30

Issue

Volume 10, Issue 3, May-June-2023

Section

Research Articles

License

Copyright (c) IJSRSET

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

[1]
Marufah Hayati, Reni Permatasari "Rainfall Prediction in Statistical Downscaling Using Tweedie Compound Response and Lasso Penalty" International Journal of Scientific Research in Science, Engineering and Technology (IJSRSET), Print ISSN : 2395-1990, Online ISSN : 2394-4099, Volume 10, Issue 3, pp.537-546, May-June-2023. Available at doi : https://doi.org/10.32628/IJSRSET23103121