Asymptotic inference for nearly nonstationary AR(1) processes (Q1099564)

The authors study the asymptotic behaviour of the least-squares estimator of the autoregressive parameter of a first order autoregression when the true value of the parameter is close to unity. By reparametrizing the parameter in the form (1-\(\gamma\) /n), where \(\gamma\) is constant, the authors derive a limiting form for the estimator as n tends to infinity. This limit involves the ratio of two stochastic integrals of standard Brownian motion, and evolves into a known form when \(\gamma =0\).