東南亞金融風暴（97-98）之時空數列分析探討

Spatial Time Series Analysis in 1997-1998 Eastern Asia Financial Crisis

溫福星(Fur-Hsing Wen) ； 陳隆麒(Long-Chi Chen) ； 吳柏林(Ber-Lin Wu)

《Asia Pacific Management Review》 5卷3期 (2000/09) Pp. 331-345

https://doi.org/10.6126/APMR.2000.5.3.06

摘要

東南亞金融風暴自1997年7月2日由泰國肇始，不到一年的時間擴展到整個東亞各個國家，也間接波及到遠在東歐的俄羅斯，整個世界金融環境嚴重受到這一波風暴擴散的影響。隨著1999年1月1日歐元的上路，探討過去一年來東南亞金融風暴的影響，除了可以瞭解金融風暴的擴散過程，亦可以預測風暴未來的變化方向，有助於未來歐元對整個世界經濟與金融環境影響的瞭解。本研究以金融地圖（financial map）與修正的時空數列模式（spatial time series model）來探討東亞七個國家：泰國、馬來西亞、菲律賓、新加坡、台灣、南韓與日本自1997年7月2日金融風暴開始至1998年9月15日對美元匯率的變化，探討其匯率變化的時間效應與空間擴散過程。研究結果發現，金融地圖可以解釋金融風暴的擴散過程與持續時間，且時空數列模式對於這七個國家匯率的變化預測要優於目前廣為使用的狀態空間（State Space model）與單變量的ARIMA模式。

關鍵字

金融風暴 ； 金融地圖 ； 匯率 ； 擴散函數 ； 時空數列模式

並列摘要

The 1997-98 East Asia financial crisis started from Thailand. Less than a year it extends not only to the whole East Asia countries but also to the Russia and East Europe, which shatters the financial environment all across the world. Since the geological orientation of this financial crisis is tentatively considered, spatial time series is a heuristic tool analysis the dynamics relationships within and between points or regions distributed across spaces. These techniques will have a profound influence on the types of spatial analyses in which financial econometrics are highly concerned.In this paper, we will use the financial map and spatial time series model to investigate the variations of exchange rate for seven countries in East Asia as well as the spread speed of financial crisis space during this period. We find that the financial map can explain spread and duration time well for this empirical studies, it is particular interesting to compare the models produced by relatively complicated during the 1997-98 East Asia financial crisis.

並列關鍵字

Financial crisis ； Financial map ； Exchange rate ； Diffusion function ； Spatial time series

參考文獻

Aroian, L. A.(1985).Time Series Analysis: Theory and Practice 6: Hydrological, Geophysical, and Spatial Applications: Proceedings of the International Conference.Amsterdam:North-Holland.

Besag, J.(1974).Spatial interaction and the statistical analysis of lattice system.Journal the Royal Statistical Society, B.36

Box, G. E. P., Jenkins, G. M.(1976).Time Series Analysis: Forecasting and Control.Oakland, CA:Holden-Day Inc..

Deutsh, S. J., Pfeifer, P. E.(1980).A three-stage iterative procedure for space-time modeling.Technometrics.22

Deutsh, S. J., Ramos, J. A.(1987).Space-time ARMA modeling of vector hydrologic sequences.Water Resource Bulletin.22(6)