時間序列分析發展至今,常常發現動態資料的走勢,隨著時間過程而演變.所以傳統的模式配適常無法得到很好的解釋,因此許多學者提出不同的模型建構方法.但是對於初始模式族的選擇,卻充滿相當的主觀與經驗認定成份.本文針對時變型時間序列分析,考慮利用知識庫,由模式庫來判斷初始模式.再藉由遺傳演算法的觀念,建立模式參數的遺傳關係.我們把這種遺傳演算法,稱之為時變遺傳演算法.針對台灣省國中數學教師人數,分別以時變遺傳演算法,狀態空間,與單變量ARIMA來建構模式,並作比較.比較結果發現,時變遺傳演算法較能掌握資料反轉的趨勢,且預測值增加較為平緩.因此時變遺傳演算法在模式建構上將是個不錯的選擇.

In time series analysis, we find often the trend of dynamic

data changingwith time. Using the traditional model fitting

can't get a good explanationfor dynamic data. Therefore, many savants developed a lot of methods formodel construction.

However, these methods are usually influenced by personal

viewpoint and experience in model base selection. In this

thesis, we discussedtime-variant time series analysis. First, we builded a model base to judge inial models by knowledge base.

Then, we set up the genetic relations of themodels' parameter. This method is called Time Variant Genetic Algorithm. We use the data if the number of junior high school mathematic teachers in Taiwan to ccompare the predictive performance of Time Variant Genetic Algorithmwith State Space and ARIMA. The forecasting performance shows the Time VariantGenetic Algorithm takes a better prediction result.

湯振鶴(1 991). 台灣區國民中學八十至八十六學年度教師需求量之推估研究報告.台灣省教育廳﹒台北市教育局、高雄市教育局合辦台中市大德國中承辦
馬信行(1987) 我國各級學校未來學生數之預測，政大學報第56期，117-147
馬信行(1 992) 我國各級學校師資之預測 政大學報第65期 63-80
詐瑞雯、吳柏林(1 994). 台灣地區國中教師數預測模式 教育與心理研究第17期29-44.
台灣省政府教育廳(1971-1994) 台灣省教育統計年報
Andel, J. ( 1993 ) . A Time Series Model with Suddenly Changing Paramenters. Journal of Time Series Analysys. vol. 14 No.2. 111-123.
Akaike, H. (1974). A New Look at The Statistical Model Identtification. IEEE Transactions on Automatic Control. AC-19. 716-723.
Box, G. E . P. and Jenkins, G. M. (1976). Time Series Analysis: Forecasting and Control. Holden-Day, San Fancisco.
Giordana, A., Saitta, L., Campidogio, M. E. and Bello, G. L. (1993). Learning Relations Using Genetic Algorithms. Advances in Artifical Intelligence. 218-229.
Goldberg, D. E. (1989). Genetic Algorithms: In Search, Optimization, and Machine Learning. Addison-Wesley Publishing Company.
Inclan, C. and Tiao, G. C. (1994). Use of Cumulative Sum of Squares for Retrospective Detection of changes of Variance. Journal of the American Statistical Association. 913-923.
Koza, 1. R. (1994) . Genetic Programming 11 : Automatic Discovery of Reusable Programs. N1IT Press, 1994.
Laurence, D. (1992). Genetic Algorithms and Financial Applications. Neural ' Genetic
and Fuzzy Systems for Chaotic Financiallvfarkets. 133-147. John Wiley and Sons Inc.
Odetayo, M. O. (1995). Knowledge Acquisition and Adaptation: A Genetic Approach. Expert Systems. vol.l2, No.1. 3-13.
Parzen, E. (1977). Multiple Time Series Modeling : Detennining The Order of Approximating Autoregressive Schemes. Multivariate Analysis IV 283-295 .
Schwartz, G. (1978). Estimating The Dimension of A Model. Ann.Statist, 6 . 461-464.
Shumway, R. H. and Stoffer, D. S. (1991). Dynamjc Linear Models With Switching.
Journal ojthe American Statistical Association. 763-769.
Wei, W. \V. S. (1990). Time Series Analysis: Univariate and Multivariate Methods. Addision-Wesley Inc.
Zurada,1. M. ( 1992). Introduction to Artifical Neural System. 56-58. West Publishing Company.