The present paper investigates the time-series properties of average rainfall and temperature readings of countries in Southeast Asia (Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, Philippines, Singapore, Thailand, Timor-Leste, and Vietnam) using monthly data, spanning from 1901 to 2016. Specifically, we analyze the time trend, persistence, and seasonality properties of the series using I(d) fractional integration framework, an approach that is found to be robust in dealing with climatological series. Antipersistence is observed in the rainfall series of Cambodia, Laos, Myanmar, and Vietnam while long memory is found in the rainfall series of the remaining countries. Thus, those countries with antipersistent rainfall are prone to erratic rainfall distribution which could lead to flooding or drought. Long memory is observed in the distribution of rainfall across all countries with increasing trend found in some cases. Both climatic variables display evidence supporting seasonality with stronger seasonality observed in Cambodia, Laos, Myanmar, Thailand, and Vietnam. Findings in the present paper are of policy relevance to climate monitoring agencies and the government.

本文使用月度数据调查了东南亚国家（文莱，柬埔寨，印度尼西亚，老挝，马来西亚，缅甸，菲律宾，新加坡，泰国，东帝汶和越南）的平均降雨量和温度读数的时间序列特性，跨度为1901年至2016年。具体而言，我们使用I（d）分数积分框架分析了该系列的时间趋势，持久性和季节性属性，该方法在处理气候系列中非常有效。在柬埔寨，老挝，缅甸和越南的降雨序列中观察到反持久性，而在其余国家的降雨序列中则发现记忆力很强。因此，那些降雨呈持久性的国家降雨分布很不稳定，可能导致洪水或干旱。在所有国家的降雨分布中观察到很长的记忆，在某些情况下呈增加趋势。这两个气候变量都显示出支持季节性的证据，在柬埔寨，老挝，缅甸，泰国和越南观察到较强的季节性。本文的发现与气候监测机构和政府的政策相关。