Loading...

Nonlinear Modelling of Return and Volatility in IRANs Auto Industry

Ebrahimi, Bababk | 2010

639 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 40685 (44)
  4. University: Sharif University of Technology
  5. Department: Management and Economics
  6. Advisor(s): Keshavarz Hadad, Golam Reza
  7. Abstract:
  8. Long memory in assets return is important on the theoretical as well as the practical aspects and since it is a special form of nonlinear dynamics, in this study an ARFIMA model was used with the aim of considering the long memory in the return conditional average of automobile industry index. Then, due to the observation of ARCH effects, the simultaneous conditional average modeling and turbulences in the automotive industry index return using an ARFIMA-FIGARCH model was considered in the residuals of the best fitted model and results were that for modeling the returns and turbulences of automobile industry stocks which has a special long memory, FIGARCH will provide better and more complete modeling. A multivariate FIGARCH model has also been used for checking the transmission of turbulence between the automobile industry index and leasing machinery and equipment. This development multivariate model is from BEKK model considering long memory parameter (d) as it will provide the long memory parameter (d) during the process of model building. The results indicate the existence of short memory and also the difference of estimated d for various models due to the models different representation of different shocks effect longevity amount on average logarithmic time series process. Results of modeling return turbulence transmission indicated the spread of automobile industry index to the leasing, machinery and equipment index. This effect in leasing worked bilaterally and it is higher from automobile index to leasing index. The Results confirmed the lead-lag effect and information flow in these two time series. Turbulence transmission from leasing stock into the machinery and production of parts and vice versa was also observed. Comparing the results of estimating the multivariate FIGARCH model with multivariate GARCH model, shows the proximity of these two models analysis structure but multivariate FIGARCH model provide more accuracy and power fitting rather than return turbulences transmission and is more accordance to the theories of economic basis.
  9. Keywords:
  10. Efficiency ; Turbulence ; Long Memeory ; Automotive Industry ; Nonlinear Modeling ; Autoregressive Fractionally Integrated Moveing Average (ARFIMA) ; General Autoregressive Conditional Heteroskedastic (GARCH)

 Digital Object List

 Bookmark