Volume 2, Issue 3, June 2014, Page: 66-70
Quantile Regression in Statistical Downscaling to Estimate Extreme Monthly Rainfall
Aji Hamim Wigena, Department of Statistics, Bogor Agricultural University, Bogor, Indonesia
Anik Djuraidah, Department of Statistics, Bogor Agricultural University, Bogor, Indonesia
Received: May 30, 2014;       Accepted: Jun. 30, 2014;       Published: Jul. 20, 2014
DOI: 10.11648/j.sjams.20140203.12      View  2955      Downloads  262
Abstract
Extreme rainfall events have been great interest in statistical downscaling. This paper concerns with developing model of statistical downscaling using quantile regression to estimate extreme monthly rainfall. Statistical downscaling relates functionally local scale response variable and global scale predictor variables. The response variable is monthly rainfall from 1979 to 2008 at station Bangkir Indonesia and the predictor variables are monthly precipitation of 64 grid of Global Circulation Model output in the same period. Principal Component Analysis is used to reduce dimension of predictors. A number of components for developing quantile regression model are determined based on Quantile Verification Skill Score. The results show that at 95th quantile the pattern of forecasted rainfall in January to December 2008 is similar to actual rainfall with correlation 0.98 and the forecasted rainfall (843 mm) in February 2008 is considered as the extreme rainfall which confirms well to the highest actual rainfall (727 mm) with probability 0.99.
Keywords
Collinear, Extreme, Principal Component Analysis, Statistical Downscaling, Quantile Regression, Logistic Regression
To cite this article
Aji Hamim Wigena, Anik Djuraidah, Quantile Regression in Statistical Downscaling to Estimate Extreme Monthly Rainfall, Science Journal of Applied Mathematics and Statistics. Vol. 2, No. 3, 2014, pp. 66-70. doi: 10.11648/j.sjams.20140203.12
Reference
[1]
IPCC. IPCC fourth assessment report: climate change 2007, Working Group I: The Physical Science Basis, (Cambridge University Press, 2007, p.996.
[2]
A.H. Wigena. Modeling of statistical downscaling using projection pursuit regression for forecasting monthly rainfall, doctoral diss., Bogor Agricultural University (in Indonesian), Indonesia. 2006.
[3]
A. Djuraidah, and A.H.Wigena. Quantile regression to explore rainfall pattern. Jurnal Ilmu Dasar, 12(1). 2011. (in Indonesian).
[4]
S.I.V. Sousa, J.C.M. Pires, F.G. Martins, M.C. Pereira, and M.C.M. Alvim-Ferraz. Potentialities of quantile regression to predict ozone concentrations, Environmetrics, 20, 2009, 147–158. (DOI: 10.1002/env.916).
[5]
J.B. Bremnes. Probabilistic forecasts of precipitation in terms of quantiles using NWP model output, Monthly Weather Review, 132, 2004, pp.338-347.
[6]
P. Friederichs, and A. Hense,. Statistical Downscaling of Extreme Precipitation Events Using Censored Quantile regression. Monthly Weather Review, 135, 2007, pp. 2365-2378.
[7]
P. Friederichs. Statistical downscaling of extreme precipitation events using extreme value theory. Extremes, 13, 2010, pp.109–132. (DOI 10.1007/s10687-010-0107-5).
[8]
I.S. Buhai. Quantile regression: overview and selected applications. Roger Koenker’s lecture notes, 2004. (the recent Netherlands Network of Economics Workshop in Groningen 2004).
[9]
S. Chamaille´-Jammesa, H. Fritz, and F. Murindagomo. Detecting climate changes of concern in highly variable environments : Quantile regressions reveal that droughts worsen in Hwange National Park, Zimbabwe. Journal of Arid Environments, 71, 2007, pp. 21–326.
[10]
T.H. Jagger and J.B. Elsner. Modeling tropical cyclone intensity with quantile regression. Int. J. Climatol, 2008, Published online in Wiley Inter Science (www.interscience.wiley.com). (DOI: 10.1002/ joc.1804).
Browse journals by subject