Marked empirical processes for non-stationary time series

Consider a first-order autoregressive process $X_i=βX_{i-1}+\varepsilon_i,$ where $\varepsilon_i=G(η_i,η_{i-1},\ldots)$ and $η_i,i\in\mathbb{Z}$ are i.i.d. random variables. Motivated by two important issues for the inference of this model, namely, the quantile inference for $H_0: β=1$, and the goodness-of-fit for the unit root model, the notion of the marked empirical process $α_n(x)=\frac{1}{n}\sum_{i=1}^ng(X_i/a_n)I(\varepsilon_i\leq x),x\in\mathbb{R}$ is investigated in this paper. Herein, $g(\cdot)$ is a continuous function on $\mathbb{R}$ and $\{a_n\}$ is a sequence of self-normalizing constants. As the innovation $\{\varepsilon_i\}$ is usually not observable, the residual marked empirical process $\hat α_n(x)=\frac{1}{n}\sum_{i=1}^ng(X_i/a_n)I(\hat{\varepsilon}_iłeq x),x\in\mathbb{R},$ is considered instead, where $\hat{\varepsilon}_i=X_i-\hatβX_{i-1}$ and $\hatβ$ is a consistent estimate of $β.$ In particular, via the martingale decomposition of stationary process and the stochastic integral result of Jakubowski (Ann. Probab. 24 (1996) 2141-2153), the limit distributions of $α_n(x)$ and $\hatα_n(x)$ are established when $\{\varepsilon_i\}$ is a short-memory process. Furthermore, by virtue of the results of Wu (Bernoulli 95 (2003) 809-831) and Ho and Hsing (Ann. Statist. 24 (1996) 992-1024) of empirical process and the integral result of Mikosch and Norvaiša (Bernoulli 6 (2000) 401-434) and Young (Acta Math. 67 (1936) 251-282), the limit distributions of $α_n(x)$ and $\hatα_n(x)$ are also derived when $\{\varepsilon_i\}$ is a long-memory process.

No reviews yet.

No comments yet.