Prediction of Logarithmic Returns of Various Risk Levels Pension Investment Funds via Box-Jenkins Models

Erdemir Ö. G., Kırkağaç M.

8th International Researchers, Statisticians and Young Statisticians Congress (8.IRSYSC), Adana, Türkiye, 28 - 30 Kasım 2024, ss.1

  • Yayın Türü: Bildiri / Özet Bildiri
  • Basıldığı Şehir: Adana
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.1
  • Hacettepe Üniversitesi Adresli: Evet

Özet

Private pension funds with increasing returns depending on the risk levels are important investment instruments that yield high returns due to their high-risk levels. In this study, the behaviors of high, medium and low risk pension investment funds in Turkey are analyzed using time series analysis methods over a six-year period. Daily logarithmic returns of various risk level pension investment funds are converted into weekly average logarithmic returns. The stationarity of the weekly average logarithmic return values of funds are examined with the help of unit root tests. Steady weekly average logarithmic return values are modeled with appropriate Box-Jenkins models or in other words Autoregressive Integrated Moving Average (ARIMA) models, a one-year forecast is compared to the real values. It has been noticed that in low risk funds, forecast values that are closer to reality and have lower errors are obtained with Box-Jenkins models.