標題: 台灣人口結構變動對產業結構之影響與預測
Application of population structure change on industry structure forecast in Taiwan
作者: 徐慶昌
Ching-chang Hsu
袁建中
Benjamin J.C. Yuan
科技管理研究所
關鍵字: 人口結構;產業結構;人口統計;產業預測;關聯性分析;迴歸預測;類神經網路;population structure;industry structure;demographic;industry forecast;relation analysis;regression model;artificial neural network
公開日期: 2005
摘要: 由於醫藥的發達與科技的進步,現代人的平均壽命也不斷的延長,而二次世界大戰後的嬰兒潮,目前也逐步邁入老年期的階段,在未來的20年內,許多國家將陸續面臨到人口結構改變的問題,台灣當然也不例外;此趨勢的產生,不但將使政府在國家政策的制定上,必需有所改變與調整,其也將進一步影響到一個國家的產業結構變化以及國家競爭力的強弱。過去人口學對產業的研究,多集中在人口移動理論的建立與環境因素的探討,而在產業結構變遷因素的探討大部份也都集中在經濟發展對產業結構的影響,較少將人口結構改變的影響也列入考慮。而本文主要則是把焦點放在人口結構的變化對產業結構所產生的影響,推估未來20年台灣人口結構的變化以及人口結構改變對各產業所產生的影響,並藉此找出在此人口結構的變化下,影響最為深遠的產業。首先,使用關聯性分析找出影響產業結構變化最為關鍵的人口結構變化因子;接著利用迴歸預測模型及人口變動要素合成法針對以上的關鍵影響因子進行預測,找出各關鍵影響因子未來20年可能的變化;最後我們將以上所得到的資料作結合,針對各產業結構指標建立相關的類神經網路模型,並帶入未來20年人口預測資料,得到各產業結構指標的預測結果。藉由所得到的結果,我們可以預測未來20來台灣產業結構變化的方向以及各產業發展的趨勢,以提供相關產業的公司制定公司的發展策略,同時也可以作為政府制定國家產業政策的一項參考依據。
Because of the improvement of medicine and technology, human life is continuously extended. The baby boomers of WWII are gradually stepping into the aged pool; many countries will be faced with the problem of population structure changes in the following years. By this trend, these countries’ governments must make some policy changes to meet the trend. Furthermore, the industry structure and national competence will also be immensely affected. Most previous researches about the impact of demographic trends are focused on the migration theory of population and the environment factors. But this research is focused on the impact of population structure change on the industry structure. We use Taiwan as example to illustrate the relation between the population structure and the industry structure. First, we use relation analysis (correlation coefficients) to build the model of the impact of population structure on industry structure. Second, we forecast the population structure in the following years by regression models and the cohort component method. Finally, we combine the relation and the population forecast of previous stage to forecast the industry structure in the following years, and to build the ANN models between the population structure and the industry structure. From the ANN models we can get the industrial development direction that related corporation can follow to make its appropriate strategy, and government can refer to formulate the national industry policies.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT009335521
http://hdl.handle.net/11536/79577
Appears in Collections:Thesis