The paper considers the estimation problem of the autoregressive parameter in the first-order autoregressive process when the noise variance is unknown. We propose a non-asymptotic technique to compensate the unknown variance, and then, to construct a point estimator with any prescribed mean square accuracy. For Gaussian noise, a fixed-width confidence interval with any prescribed coverage accuracy is proposed. The results of simulations correspond to the theoretical results.
Международная научная конференция "Робастная статистика и финансовая математика – 2020" (15-16 декабря 2020 г.) : сборник статей. Томск, 2021. С. 56-61