PERAMALAN CURAH HUJAN KOTA BANDUNG MENGGUNAKAN MODEL FUNGSI TRANSFER MULTIVARIAT PADA DERET BERKALA MUSIMAN

• Main Author: Prahesti, Danica Dwi (Author)
• Format: Book
• Published: 2016-06-28.
• Subjects: Thesis; NonPeerReviewed
• Online Access: Link Metadata

Summary

Model fungsi transfer pada deret berkala pola musiman merupakan penggabungan karakteristik dari model analisis regresi berganda dengan karakteristik model deret berkala pola musiman SARIMA (Seasonal Autoregresive Integrated Moving Average). Model fungsi transfer yang memiliki lebih dari dua deret berkala pada variabel input disebut dengan model fungsi transfer multivariat. Prosedur untuk menentukan model fungsi transfer diawali dengan menghilangkan pengaruh musiman (deseasonalized) pada semua deret berkala, kemudian untuk menentukan model fungsi transfer multivariat dimulai dengan pembentukan model fungsi transfer univariat pada masing-masing variabel input, kemudian dilakukan pembentukan secara serentak menjadi model fungsi transfer multivariat. Model fungsi transfer pada deret berkala pola musiman diterapkan untuk meramalkan jumlah curah hujan di kota Bandung, dengan variabel input berupa data kelembaban udara, suhu udara, tekanan udara, dan kecepatan angin. Data yang digunakan dimulai dari bulan januari 2000 hingga desember 2013. Hasil penelitian ini menunjukkan bahwa : (1) model peramalan curah hujan menggunakan model fungsi transfer multivariat pada deret berkala musiman yaitu: y_t=0,782085(kelembaban)_t+0,232096(suhu)_(t-1)+1,014625(tekanan)_(t-1)+0,370232(kecepatan angin)_(t-1)+(1+0,548733B)(1+0,927859B^12 ) a_t ; (2) hasil peramalan curah hujan kota bandung untuk 12 bulan ke depan menunjukkan terjadi fluktuasi dimana jumlah curah hujan paling tinggi terjadi pada bulan februari 2014 dan jumlah curah hujan terendah pada bulan januari 2014. Kata Kunci : Deret Berkala, SARIMA, Fungsi Transfer, Multivariat, Curah Transfer function models for time series seasonal pattern which is combination characteristic of the multiple regression analysis model with a characteristics of the time series seasonal pattern models SARIMA (Seasonal Autoregresive Integrated Moving Average). Transfer function models has more than two time series on the input variables is called multivariate transfer function model. The procedure for determining the transfer function models begins with remove the influence of seasonal (deseasonalized) at all time series, then for determine multivariate transfer function models begins with the establishment of the univariate transfer function models at each input variable, and then do the establishment simultaneously be a multivariate transfer function models. multivariate transfer function models for time series seasonal pattern applied to predict the amount of rainfall at Bandung city, with a form of data input variable is air humidity, air temperature, air pressure, and wind speed. Used data starts from January 2000 to December 2013. Results from this study indicate that : (1) forecasting models of rainfall used transfer function models that : y_t=0,782085(humidity)_t+0,232096(temperature)_(t-1) +1,014625(pressure)_(t-1)+0,370232(wind speed)_(t-1)+(1+0,548733B) (1+0,927859B^12 ) a_t; (2) the results of forecasting for rainfall Bandung city for 12th month ahead showed fluctuation where the number of the rainfall highest uccurred on february 2014 and the number of rainfall lowest on january 2014. Key word : time series, SARIMA, transfer function, multivariate, rainfall

Item Description

http://repository.upi.edu/29769/1/S_MAT_1104728_Title.pdf
http://repository.upi.edu/29769/2/S_MAT_1104728_Abstract.pdf
http://repository.upi.edu/29769/3/S_MAT_1104728_Table_of_content.pdf
http://repository.upi.edu/29769/4/S_MAT_1104728_Chapter1.pdf
http://repository.upi.edu/29769/5/S_MAT_1104728_Chapter2.pdf
http://repository.upi.edu/29769/6/S_MAT_1104728_Chapter3.pdf
http://repository.upi.edu/29769/7/S_MAT_1104728_Chapter4.pdf
http://repository.upi.edu/29769/8/S_MAT_1104728_Chapter5.pdf
http://repository.upi.edu/29769/9/S_MAT_1104728_Bibliography.pdf
http://repository.upi.edu/29769/10/S_MAT_1104728_Appendix.pdf