Statistical Downscaling with Bayesian Quantile Regression to Estimate Extreme Rainfall in West Java

  • Eko Primadi Hendri Department of Statistics, IPB University, Bogor, 16680, Indonesia
  • Aji Hamim Wigena Department of Statistics, IPB University, Bogor, 16680, Indonesia
  • Anik Djuraidah Department of Statistics, IPB University, Bogor, 16680, Indonesia
Keywords: Statistical Downscaling, Bayesian Quantile Regression, LASSO, MCMC.

Abstract

West Java is one of the largest regions producing rice in Indonesia. Information on rainfall is very important for farmers to anticipate extreme events that can cause losses in agriculture. Extreme rainfall patterns can be modeled by Bayesian quantile regression. Parameters of the model are estimated by MCMC. This study used statistical downscaling to obtain relationship between global scale data and local scale data. The data were monthly rainfall data in West Java based on three type of land, low, medium, high land, and GCM output data. LASSO regularization was used to overcome multicollinearity problem in GCM output data. The purpose of this study was to model Bayesian quantile regression in each type of land. The Bayesian quantile regression model in the lowlands can predict extreme rainfall accurately and consistently in one year ahead.

References

. Badan Metorologi, Klimatologi dan Geofisika (BMKG). Laporan Meteorologi, Klimatologi, dan Geofisika : Jakarta, 2008

. Badan Pusat Statitik. Ringkasan Eksekutif, Luas Panen dan Produksi Padi di Indonesia 2018 : Jakarta, 2018.

. Cahyani, T.B.N. “Statistical Downscaling Modelling with Ridge and Elastic-net Regularized Quantile Regression for Rainfall Prediction in Indramayu.” M.Sc. Thesis, IPB University, Bogor, 2016.

. Chen. C and Wei. Y. “Computation Issues For Quantile Regression.” The Indian Journal Statistics, vol. 67(2), pp. 399-417, 2005.

. Djuraidah. A. and Wigena. A. H. “Regresi kuantil untuk Eksplorasi Pola Curah Hujan di Kabupaten Indramayu”. Jurnal Ilmu Dasar. vol. 12(1), pp. 50-56, 2011.

. Koenker. R. and Bassett. G. “Regression Quantile”. Econometrica. vol. 46(1), pp. 33-50, 1987.

. Kozumi. H. and Kobayashi. G. “Gibbs Sampling Methods for Bayesian Quantile Regression”. Journal of Statistical Computation and Simulation. vol. 81(11), pp. 1565-1578, 2011.

. Geman. S. and Geman. D. “Stochastic relation, gibbs distribution, and the bayesian restoration of image”. IEEE Transaction on Pattern Analisys and Machine Intelligence. vol. 6, pp. 721-741. 1984.

. Mondiana. Y.Q. “Statictical Downscaling Modeling with Quantile Regression to Estimate Extreme Precipitation (A Case Study in Bangkir Station, Indramayu).” M.Sc. Thesis, IPB University, Bogor, 2012.

. Nuryanto. B. et al. “Pengaruh Tinggi Tempat dan Tipe Tanaman Padi Terhadap Keparahan Penyakit Hewan Pelepak”. Jurnal. Penelitian Pertanian Taman Pangan. vol. 33(1), 2014.

. Rosyidie. A. “Banjir: Fakta dan Dampaknya, Serta Pengaruh Guna Lahan. Journal Perencanaan Wilayah dan Kota”. vol. 24(3), pp. 241-249, 2013.

. Santri. D. “Statistical Downscaling Modeling with Quantile Regression using LASSO to Estimate Extreme Rainfall.” M.Sc. Thesis, IPB University, Bogor, 2016.

. Suciantini. “Interaksi Iklim (Curah Hujan) Terhadap Produksi Tanaman Pangan di Kabupaten Pacitan,” in Proc. Seminar Nasional Masyarakat Biodiversitas Indonesia, 2015, pp. 358-365.

. Pribadi. H.Y. “Variabilitas Curah Hujan dan Pergeseran Musim di Wilayah Banten Sehubungan dengan Variasi suhu Muka Laut Perairan Indonesia, Samudra Pasifik dan Samudra Hindia.” M.Sc. Thesis, Universitas Indonesia, Depok, 2012.

. Wigena. A.H. “Statistical Downscaling Modeling using projection Pursuit Regression to Forecash Monthly Rainfall.” Dr. dissertation, IPB University, Bogor, 2006.

. Yu. K. and Moyeed. R.A. “Bayesian Quantile Regresion”. Statistics & Probability Latter. vol. 54(4), pp. 437-447, 2001.

. Zakarina. H. “Lasso and Ridge Quantile Regression using Cross Validation to Estimate Extreme Rainfall.” M.Sc. Thesis, IPB University, Bogor, 2016.

Published
2019-07-23
Section
Articles