𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Limit theory for random coefficient first-order autoregressive process under martingale difference error sequence

✍ Scribed by Zhi-Wen Zhao; De-Hui Wang; Yong Zhang


Publisher
Elsevier Science
Year
2011
Tongue
English
Weight
229 KB
Volume
235
Category
Article
ISSN
0377-0427

No coin nor oath required. For personal study only.

✦ Synopsis


Phillips and Magdalinos (2007)

[1] gave the asymptotic theory for autoregressive time series with a root of the form ρ n = 1 + c/k n , where k n is a deterministic sequence. In this paper, an extension to the more general case where the coefficients of an AR(1) model is a random variable and the error sequence is a sequence of martingale differences is discussed. A conditional least squares estimator of the autoregressive coefficient is derived and shown to be asymptotically normal. This extends the result of Phillips and Magdalinos (2007) [1] for stationary and near-stationary cases.