Advances and Applications in Statistics
Volume 24, Issue 1, Pages 49 - 66
(September 2011)
|
|
PARAMETER ESTIMATION IN A SYSTEM OF STOCHASTIC DIFFERENCE EQUATIONS
N. Josephy, M. Predescu and S. Woolford
|
Abstract: We reformulate a nonlinear difference equation model by Wikan and Eide [11] by incorporating stochastic errors as suggested by Dennis et al. [4]. Using simulated data, we explore the performance of maximum likelihood estimation to estimate both model parameters and error term parameters for the model. We find that the performance of the estimation is strongly influenced by the different stability regions associated with the deterministic model. Additionally, the accuracy of the estimation of the model parameters can depend upon the parameters of the error distribution. We also evaluate the effect of sample size and error term distribution on estimation accuracy. |
Keywords and phrases: parameter estimation, stochasticity, difference equations. |
|
Number of Downloads: 338 | Number of Views: 1417 |
|