<![CDATA[This research was conducted to study and know the characteristics of periodic and stochastic models of daily rainfall in the Way Jepara. This study used secondary data, namely daily rainfall data with a data length of 13 years (1997-2013) from the station Braja Arjosari, Braja Indah, and Jepara Lama. In this study , daily rainfall data is processed into the form of time series before it was trans- formed into the spectrum of rainfall using the program FFT (Fast Fourier Transform). One year of daily rainfall data periodicity, only used 512 rainfall data are periodic. Spectrum rainfall pro- duce data series stochastic rainfall is assumed as the difference (error) between the actual rainfall data with periodic rainfall models. Based on data from the series of stochastic, the stochastic com- ponent is calculated using the approach of autoregressive models. Stochastic model generated us- ing the least squares method (least square method) or autoregressive model of order three. Vali- dation of stochastic series between the measured data and the model is done by calculating the correlation coefficient. From the research, the correlation coefficient obtained an average of three stations of the daily rainfall. Value of the average correlation coefficient ( R ) between the data with a model pe- riodic rainfall amounted to 0.97305, the data series is stochastic and stochastic models of 0.99150, and between data and periodic stochastic models are at 0.99963. From the results, it can be concluded that the periodic stochastic models of rainfall Way Jepara region gives very accurate results and the approach significantly.]]>
Copyrights © 2015
