PEMODELAN PERIODIK DAN STOKASTIK CURAH HUJAN KOTA BANDAR LAMPUNG
ABSTRACT: This research aims
to study the periodic and stochastic modeling of daily precipitation data
series. The study was carried out using data of daily rainfall data with a
length of 13 years (1987 2000)from Pahoman, Sumur Putri, and Sumber Rejo rainfall
stations. The rainfall series used is as-sumed to be free from the influence of
the trend. This research use how to convert data into timeseries sepektrum rain
precipitation using program FFT (Fast Fourier Transform). Periodicity of daily
rainfall data is presented using a periodic component 512. Stochastic rainfall
series of rain-fall data is assumed as the difference (error) between the
rainfall data with periodic rainfall model using 512 components. Based on the
data series on stochastic stochastic component, calculatedusing the approach of
the autoregressive model. Stochastic models presented by using a autore- gressive
model of order two.
The results of this research is the value of the correlation coefficient
of the average precipitation three stations. For this study, the average
correlation coefficient (R) between data and model of0,9719 is periodic,
between data and stochastic model is 0,9974, and between the data and the model
with stochastic is 0,9974. From these results it can be concluded that periodic
stochasticmodel of the rainfall of the Bandar Lampung City that processing
model using 512-series components, presented a very significant approach.
Penulis: Ahmad Zakaria,
Kartini Susilowati, Rasimin
Kode Jurnal: jptsipildd140782
