Some Recent Developments in Time Series Analysis, Correspondent Paper
这篇通讯论文综述了时间序列分析中五个领域的近期进展，包括结构模型、单位根、模型选择、自回归条件异方差和稳健估计，适合希望快速了解该领域前沿的经济学和统计学研究者。
As was the case when preparing my previous survey (Newbold, 1984), the sheer volume of recent published material in time series analysis, together with its diversity, dictated that attention here should be restricted to a few selected topics. In making this choice, I have tried both to cover a broad range of developments and to highlight those areas which have seen the most exciting progress and which promise further growth of attention in the near future. No doubt this selection will not be to the taste of all readers, and inevitably to some extent it reflects my own prejudices. In the following sections we consider recent advances in structural and unobserved components models, models with unit autoregressive roots, model selection and order estimation, autoregressive conditional heteroscedasticity, and robust estimation of time series models. It must be admitted that the problem of model selection has received attention in both my previous reviews. Nevertheless, the quantity and quality of recent work on this topic dictate further discussion here.