Prediction Interval in Seasonal Autoregressive Integrated Moving Average (SARIMA) Model for Rainfall Forecasting and Drought

Vita Mami Nikmatillah, Dian Anggraeni, Alfian Futuhal Hadi

2018

Abstract

The prediction interval in the forecasting process is the most important part of knowing indication of uncertainty in forecasts value. The uncertainty also serves to reduce the forecasting errors that occur. This research uses SARIMA model in rainfall forecasting process in Jember Regency. In addition to calculating predictive interval values, the predicted values ​​generated by the SARIMA model are transformed in Standardized Precipitation Index (SPI) values ​​to determine the classification of drought levels. The results showed that the predictive interval value is presented in the form of upper limit and lower limit of precipitation value of rainfall. The resulting drought prediction interval indicates that droughts in the four regions of Jember Regency due to the lower limit value reaching minus. The drought index at the SPI value shows almost all areas in normal conditions. However, Zone 1 in January and Zone 2 in November contained a moderately wet month where the rain intensity was greater was greater rather than the other months in the same zone. The difference in classification results from predictive intervals and the SPI method is very large. The predicted underestimate prediction value indicates that the prediction interval is poor in interpreting a region's drought.

Paper Citation

in Harvard Style

Nikmatillah V., Anggraeni D. and Hadi A. (2018). Prediction Interval in Seasonal Autoregressive Integrated Moving Average (SARIMA) Model for Rainfall Forecasting and Drought.In Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs, ISBN 978-989-758-407-7, pages 101-107. DOI: 10.5220/0008517801010107

in Bibtex Style

@conference{icmis18,
author={Vita Mami Nikmatillah and Dian Anggraeni and Alfian Futuhal Hadi},
title={Prediction Interval in Seasonal Autoregressive Integrated Moving Average (SARIMA) Model for Rainfall Forecasting and Drought},
booktitle={Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs,},
year={2018},
pages={101-107},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008517801010107},
isbn={978-989-758-407-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs,
TI - Prediction Interval in Seasonal Autoregressive Integrated Moving Average (SARIMA) Model for Rainfall Forecasting and Drought
SN - 978-989-758-407-7
AU - Nikmatillah V.
AU - Anggraeni D.
AU - Hadi A.
PY - 2018
SP - 101
EP - 107
DO - 10.5220/0008517801010107