scholarly journals PERAMALAN JUMLAH PENDERITA DEMAMBER DARAH DENGUE DI KABUPATEN JOMBANG JAWA TIMUR DENGAN PENDEKATAN FUNGSI TRANSFER SINGLE INPUT

2018 ◽  
Vol 15 (2) ◽  
pp. 10
Author(s):  
Sediono Sediono

AbstractForecasting  is an important things in time series analysis, because by obtaining a convenient model that is statictically appropriate. Clearly, that can be used to predict the structure of future data form. Transfer function is one of mathematical model  in time series analysis, that can be used to forecasting time index data both univariate and multivariate. Transfer function describes the predictive value  of the output series (Yt) based on the value of one or more input series(Xt). The single input transfer function model is a transfer function model that uses one variable as input series (Xt), where each series of both input series and output series must be a stationary time series model, both stationary in the mean and stationary in variant. One of the used transfer function is to govern a model and forecasting of the number of cases dengue fever (Yt) in Kabupaten Jombang, East Java, where the input variable based on data of rainfall (Xt). From the result of this study was obtained that model of transfer function has a equation Y𝑡 = 0,0542X𝑡+  (1 − 0,7309𝐵)(1 + 0,6568𝐵12) with parameter ωo = 0.0542, ∅1 = 0.7309 and  Φ12 = -0.6568. From the model, it can be interpreted that the number of dengue sufferers for a particular month was influenced by the rainfall on those month and the months before. According to the model of the transfer function, it can be used to forecast the number of sufferers of dengue fever in Kabupaten Jombang  for period next 20 months. After compared between data of forecasting and actual data, there exists equally  trend, namely 15 months of 20 month that are forecasted, such that  it can be explain that majority 75% of the results of forecasting in this study are valid. Keywords: forecasting , single input transfer function, stationer point, Dengue fever  Abstrak Peramalan adalah sesuatu hal yang penting dalam analisis runtun waktu, karena dengan diperolehnya sebuah model yang tepat secara statistik, jelas hal tersebut dapat digunakan untuk memprediksi struktur pola data yang akan datang. Fungsi transfer merupakan salah satu model matematis dalam analisis runtun waktu  yang dapat digunakan untuk  peramalan data indekswaktu baik univariat maupun multivariat. Fungsi transfer menggambarkan nilai prediksi  dari  output series (Yt) berdasarkan nilai satu atau lebih input series (Xt). Model fungsi transfer  single input  adalah model fungsi transfer yang menggunakan satu variabel sebagai input series (Xt), dimana masing-masing series baik input series maupun output series keduanya harus sama-sama merupakan model runtun waktu yang stasioner, baik stasioner dalam mean maupun stasioner dalam varian. Salah satu penggunaan  model fungsi transfer ini adalah untuk pembuatan model dan peramalan jumlah kasus demam berdarah dengue (Yt)  di Kabupaten Jombang Jawa Timur, dengan variabel inputnya berdasarkan data curah hujan (Xt). Dari hasil penelitian diperoleh  model fungsi transfer yang memiliki persamaan  Y𝑡 = 0,0542X𝑡 +  (1 − 0,7309𝐵)(1 + 0,6568𝐵12) 𝑎𝑡 , dengan parameter ωo = 0,0542, ∅1 = 0,7309, dan Φ12 = -0,6568. Dari model tersebut dapat diinterpretasikan bahwa jumlah penderita demam berdarah dengue pada suatu bulan dipengaruhi curah hujan pada bulan itu, dan dipengaruhi oleh beberapa gangguan pada bulan-bulan sebelumnya. Selanjutnya berdasarkan model fungsi transfer tersebut dapat digunakan untuk peramalan jumlah penderita demam berdarah dengue di Kabupaten Jombang untuk periode 20 bulan kedepan. Setelah dilakukan perbandingan antara data hasil peramalan dengan data aktual, terdapat kesamaan trend yaitu sejumlah 15 bulan dari 20 bulan yang diramalkan, sehingga dapat dijelaskan bahwa sebagian besar yaitu 75% dari hasil peramalan  dalam penelitian ini adalah valid. Kata Kunci : Peramalan, Fungsi transfer single input, stasioner, Demam Berdarah Dengue. 

1984 ◽  
Vol 21 (01) ◽  
pp. 88-97
Author(s):  
Victor Solo

The consistency is developed under mild conditions for the least squares estimator of the parameters of a transfer function time series model.


2015 ◽  
Vol 23 (2) ◽  
pp. 30-36 ◽  
Author(s):  
Patrik Sleziak ◽  
Kamila Hlavčová ◽  
Ján Szolgay

Abstract The paper presents an analysis of changes in the structure of the average annual discharges, average annual air temperature, and average annual precipitation time series in Slovakia. Three time series with lengths of observation from 1961 to 2006 were analyzed. An introduction to spectral analysis with Fourier analysis (FA) is given. This method is used to determine significant periods of a time series. Later in this article a description of a wavelet transform (WT) is reviewed. This method is able to work with non-stationary time series and detect when significant periods are presented. Subsequently, models for the detection of potential changes in the structure of the time series analyzed were created with the aim of capturing changes in the cyclical components and the multiannual variability of the time series selected for Slovakia. Finally, some of the comparisons of the time series analyzed are discussed. The aim of the paper is to show the advantages of time series analysis using WT compared with FT. The results were processed in the R software environment.


Sign in / Sign up

Export Citation Format

Share Document