AI Summary & Key Takeaways
AI InsightAs a result, it is possible to say that determination of forecasting methods firstly according to the components of time series will contribute positively to decision makers in terms of both time and calculation
📌 3 Key Takeaways
Abstract
Although there are many forecasting methods in the literature, the most basic ones are moving averages, exponential smoothing methods, advanced exponential smoothing methods and Pegel’s classification method. It is difficult and time-consuming to calculate for all of these basic methods, which consist of a total of 18 different methods. Thus, in this paper, the trend and seasonal components of BIST-100 index between 2005 and 2017 as annually and between January 2010 and April 2018 as monthly average were analyzed and forcested based on 6 of these basic forecasting methods. As a result, it is possible to say that determination of forecasting methods firstly according to the components of time series will contribute positively to decision makers in terms of both time and calculation
Özet
Literatürde birçok tahmin tekni÷i yer almasÕna ra÷men en temel tahmin teknikleri hareketli ortalamalar, üstel düzeltme, ileri düzey üstel düzeltme tahmin teknikleri ve Pegels’in sÕnÕflandÕrma yöntemidir. Toplamda 18 farklÕ yöntemden oluúan bu temel tekniklerin hepsi için hesaplamalar yapmak oldukça zor ve zaman alÕcÕdÕr. Bu nedenle, çalÕúmada 2005 – 2017 arasÕndaki yÕllÕk ve 2010 Ocak – 2018 Nisan arasÕndaki aylÕk BøST-100 tüm endeks ortalama de÷erlerinin trend ve mevsimlik bileúenleri incelenmiú ve temel tahmin tekniklerinden 6 tanesine göre tahminler yapÕlmÕútÕr. Sonuç olarak zaman serisinin bileúenlerine göre öncelikle tahmin tekniklerinin belirlenmesinin hem zaman olarak hem de hesaplama açÕsÕndan karar vericilere olumlu katkÕ sa÷layaca÷ÕnÕ söylemek mümkündür.