PENENTUAN FAKTOR-FAKTOR YANG MEMPENGARUHI INTENSITAS CURAH HUJAN DENGAN ANALISIS DISKRIMINAN GANDA DAN REGRESI LOGISTIK MULTINOMIAL (Studi Kasus: Data Curah Hujan Kota Semarang dari Stasiun Meteorologi Maritim Tanjung Emas Periode Oktober 2018 – Maret 2019)

*Shella Faiz Rohmana  -  Departemen Statistika FSM Universitas Diponegoro, Indonesia
Agus Rusgiyono  -  3Departemen Statistika FSM Universitas Diponegoro, Indonesia
Sugito Sugito  -  , Indonesia
Received: 13 Feb 2020; Published: 18 Feb 2020.
View
Open Access Copyright 2020 Jurnal Gaussian
License URL: http://creativecommons.org/licenses/by-nc-sa/4.0

Citation Format:
Abstract

Meteorologist develop rainfall forecasting methods to obtain better and more accurate rainfall information. One of them is the research of grid data and the method of grouping rainfall. According to BMKG, rainfall is classified into light, medium, and heavy rain. This study aims to determine the factors that influencing rainfall grouping using multiple discriminant analysis with a stepwise selection method. This study uses the daily climate data of Semarang City for period of October 2018 to March 2019. Based on its partial F value, the wind speed variable is eliminated so the significant variable on rainfall grouping are air temperature, air humidity, and wind direction. This analysis produces discriminant scores obtained from linear combinations between discriminant weights and observation values of significant independent variable. The classification procedure is based on the discriminant score each observations compared to cutting score resulted in classification accuracy of 62.89%. Multinomial logistic regression analysis is used to determine the effect of independent variables on rainfall intensity using the odds ratio. This analysis produces an estimate of the conditional probability of each group using significant independent variables are air temperature, air humidity, wind speed, and wind direction. The classification procedure is based on the largest conditional probability value between rainfall groups resulted in classification accuracy of 69.80%.

 

Keywords: multiple discriminant analysis, multinomial logistic regresion, classification accuracy, rainfall

Keywords: multiple discriminant analysis, multinomial logistic regresion, classification accuracy, rainfall

Article Metrics: