Endüstri Mühendisliği Bölümü Yayın Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/279
Browse
Browsing Endüstri Mühendisliği Bölümü Yayın Koleksiyonu by Author "2337"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Conference Object Citation Count: Bayrak, O.T.; Akkaya, A.D.,"Autoregressive Models With Stochastic Design Variables and Nonnormal İnnovations",International Conference On Applied Mathematics, Simulation, Modelling - Proceedings, (2011).Autoregressive Models With Stochastic Design Variables and Nonnormal İnnovations(2011) Türker Bayrak, Özlem; Dener Akkaya, Ayşen; 56416; 2337In autoregression models the design variable has traditionally been assumed to be non-stochastic and innovations are normal. In most real life situations, however, the design variable is stochastic having a non-normal distribution as the innovations. Modified maximum likelihood method is utilized to estimate unknown parameters in such situations. Closed form estimators are obtained and shown to be efficient and robust.Article Citation Count: Türker Bayrak, Ö., Akkaya, A.D. (2010). Estimating parameters of a multiple aoutoregressive model by the modified maximum likelihood method. Journal of Computational and Applied Mathematics, 233(8), 1763-1772. http://dx.doi.org/10.1016/j.cam.2009.09.013Estimating parameters of a multiple aoutoregressive model by the modified maximum likelihood method(Elsevier Science, 2010) Türker Bayrak, Özlem; Dener Akkaya, Ayşen; 56416; 2337Abstract: We consider a multiple autoregressive model with non-normal error distributions, the latter being more prevalent in practice than the usually assumed normal distribution. Since the maximum likelihood equations have convergence problems (Puthenpura and Sinha, 1986) , we work out modified maximum likelihood equations by expressing the maximum likelihood equations in terms of ordered residuals and linearizing intractable nonlinear functions (Tiku and Suresh, 1992) . The solutions, called modified maximum estimators, are explicit functions of sample observations and therefore easy to compute. They are under some very general regularity conditions asymptotically unbiased and efficient (Vaughan and Tiku, 2000) . We show that for small sample sizes, they have negligible bias and are considerably more efficient than the traditional least squares estimators. We show that our estimators are robust to plausible deviations from an assumed distribution and are therefore enormously advantageous as compared to the least squares estimators. We give a real life example
