Bu çal ışma, Çorum Çat Deresinde ölçülen y ı ll ı k ak ı m serisinin modellenmesi amac ı yla yap ı lm ışt ı r. Yı ll ı k ak ı m serisinin modellenmesinde ARMA p,q modelleri kullan ı lm ışt ı r. Uygun modelin seçiminde, ARMA p,q modellerinden hesaplanan kal ı nt ı ları n bağı ms ı z olup olmad ı klar ı göz önüne al ı nmışt ı r. Yı ll ı k ak ı m serisi için, ARMA 2,2 , ARMA 1,1 , ARMA 3,1 , ARMA 2,2 , ARMA 1,3 ve ARMA 3,0 modelleri uygun bulunmu ştur. Seçilen modellerden tahmin edilen ak ı mlar ı n, gözlenen ak ı mlardan farklar ı n ı n karelerinin toplam ı saptanmışt ı r. Bu toplamlar ı n en küçük değeri ARMA 1,3 modelinden elde edilmi ştir. Bundan dolay ı Çorum Çat Deresi y ı ll ı k ak ı m serisinin tahmininde ARMA 1,3 modeli daha uygun bulunmu ştur
This study was made to model annual flow series measured on Çorum Cat Stream. ARMA p,q models were used in modeling annual flow series. In selecting appropriate model, It was taken into account whether the residuals calculated from ARMA p,q models were independent. For annual flow series, ARMA 2,2 , ARMA 1,1 , ARMA 3,1 , ARMA 2,2 , ARMA 1,3 and ARMA 3,0 models were appropriate. The sum square differences were extracted for the observed flows and the flows forecasted from the selected models. The lowest value of these sum square differences was obtained from ARMA 1,3 model. Therefore, ARMA 1,3 model was more suitable in forecasting annual flow series of Çorum Cat Stream.
Primary Language | Turkish |
---|---|
Journal Section | Research Article |
Authors | |
Publication Date | January 1, 2003 |
Submission Date | January 1, 2003 |
Published in Issue | Year 2003 Volume: 09 Issue: 01 |
Journal of Agricultural Sciences is published open access journal. All articles are published under the terms of the Creative Commons Attribution License (CC BY).