MODELING ISE100 WITH CONTINUOUS AR(1) MODEL

Abstract

Great majority of the studies on Istanbul Stock Exchange market (ISE100) have focused on various type of discrete modeling such as AR/MA, ARIMA, GARCH, Vector AR and extensions of GARCH modeling. The importance of finding a suitable model for a stock exchange market and having an efficient forecast results from the model is undisputable. In this study we will model ISE100 with simple AR(1) model and taking a step further in analysis to continuous modeling. Recent challenge in financial time series modeling is to find an appropriate continuous model for the data used. In our case continuous AR(1) (CAR(1)) model will be applied to ISE100 and the results of the financial modeling will be evaluated.

Keywords

• ISE100, Continuous Modeling, CAR(1), Discrete Modeling, Ornstein-Uhlenbeck Process

References

1. Box, G.E.P. and G.M. Jenkins (1970), Time series analysis: Forecasting and control, San Francisco: Holden-Day.
2. Brockwell, P. J. (1995), ‘A note on the embedding of discrete-time ARMA processes’, Journal of Time Series Analysis 16(5), 451–460.
3. Brockwell, P. J. (2000), Heavy-tailed and non-linear continuous-time ARMA models for financial time series, University of Hong Kong: Centre of Financial Time Series, Imperial College Press, London, pp. 3–23.
4. Jones, R. H. (1981), Fitting a continuous time autoregression to discrete data. In Applied Time Series Analysis //(Ed.. D. F. Findley), pp. 651-682. Academic Press, New York.
5. Jones, R. H. (1985), Time series analysis with unequally spaced data. In Time Series in the Time Domain, Handbook of Statistics 5 (Eds., E. J. Hannan, P. R. Krishnaiah and M. M. Rao), pp. 157— 178, North Holland, Amsterdam.
6. Phillips A.W. (1959), The estimation of parameters in systems of stochastic differential equations, Biometrika 46,67-76.