中文摘要：

针对湖南省自然灾害经济损失发生的现状，结合傅里叶级数的相位特征，利用傅里叶级数时间序列模型给出自然灾害损失的变化趋势。根据1986--2007年湖南省自然灾害所造成的直接经济损失数据，在SPSS环境下进行拟合，并对未来5年湖南省的自然灾害所造成的直接经济损失进行预测。结果表明：傅里叶级数时间序列模型符合自然灾害损失发生的特点，有效地解决了传统损失预测模型在灾害经济损预测中误差大、稳定性差的缺陷，提高了预测精度；且得出未来5年湖南灾害灾情仍然具有递增的趋势的结论。采用该模型对自然灾害损失进行预测，其结果比较可靠，可供减灾决策部门参考。

英文摘要：

With the situation of natural disaster economic loss in Hunan province as an example, and combining the phase characteristics of Fourier Series, the changing tendency of natural disaster loss was given by using Fourier series time series model. According to the data of direct economic loss from 1986 to 2007 in Hunan province caused by natural disasters, a fitting was carried out under the SPSS environment, and the direct economic loss of Hunan province caused by natural disasters in future 5 years was predicted. The results show that the Fourier series time series model is highly in accord with the occurring features of natural disasters, effectively solves the shortcomings of traditional loss prediction model such as big errors and bad stability, and improves the accuracy of prediction. In addition, it also shows that the disaster situation in Hunan province in future 5 years still has a tendency to increase. This study can provide reference for natural disaster reduction departments.