研究生: |
王舜平 SHUN-PING WANG |
---|---|
論文名稱: |
景氣對策信號對投資型保險顧客購買行為影響之實證研究 A Study on Buying Behavior of Investment Insurance Influenced by the Monitoring Indicators |
指導教授: |
徐俊傑
Chiun-Chieh Hsu |
口試委員: |
黃世禎
Shih-Chen Huang 謝樹明 none |
學位類別: |
碩士 Master |
系所名稱: |
管理學院 - 資訊管理系 Department of Information Management |
論文出版年: | 2009 |
畢業學年度: | 97 |
語文別: | 中文 |
論文頁數: | 103 |
中文關鍵詞: | 投資型保險 、資料探勘 、購買預測 、景氣信號 |
外文關鍵詞: | Investment Insurance, Data Mining, Purchase Forecast, Monitoring Indicators |
相關次數: | 點閱:184 下載:3 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
2001年開發出兼顧顧客投資與保險雙重需求的投資型保險,在2007年投資型保險商品總保費達6,018億,但是在投資型保險亮眼的銷售成績後,2007年8月也開始浮現金融海嘯的危機,2009年1-3月業績更較2008年同期下滑90.4% ,這個現象顯示出市場景氣變化嚴重影響投資型商品的銷售。然而目前投資型保險相關領域之資料探勘研究甚少,更無真正全面應用各地區實際銷售資料利用資料探勘技術進行購買預測的相關文獻。
資料探勘是解決此問題的好方法,然而目前投資型保險相關領域之資料探勘研究甚少,更無真正全面應用各地區實際銷售資料利用資料探勘技術進行購買預測的相關文獻。目前投資型商品銷售策略多為各公司決策階層由過去經驗、通路需求或市場觀察而決定,欠缺一個可依景氣變化調整銷售策略之預測模型可供決策參考。萬一商品策略錯誤,銷售不如預期則將浪費開發與銷售準備所投入的大量資源。
因此本研究以資料探勘之決策樹技術,某保險公司2007年10月~2009年3月之投資型保險交易資料,從中取得保戶人文變數,保單變數,結合行政院經建會之景氣燈號變數,以決策樹之資料探勘技術,依險種類別分別建構五種不同投資型保險商品之銷售預測模型,以作為未來保險業界可依照景氣變化調整銷售商品及目標客群之參考。根據實驗數據顯示五種投資型商品之預測模型正確率皆達到85%以上,可供未來業者將來制定銷售策略,掌握最佳銷售時機與目標客戶之重要決策參考。
In 2001, an investment insurance considering both investment and insurance demands of customers was proposed in the market, which made sales of investment insurance reached 6,018 billion in total premiums in 2007. However, investment insurance after brisk sales performance began to surface financial tsunami crisis in August 2007. It resulted in that the insurance sale from January to March in 2009 fell 90.4% compared with that in the same period of 2008. The phenomenon revealed that the change of market economy can deeply influence the sales of investment insurance.
Although data mining is a good approach of solving this problem, it is difficult to find the related researches about investment insurance using data mining research, let along using actual sale data in data mining to solve this problem. The current sale strategy for the investment insurance product only depends on the past experience of the company, the demand of sales channel, and observations on market. No prediction model is used for decision-making and strategy adjustment according to economy deviation. If sales strategy were inproper, sales would fall below expectation and lead to waste of resource for product development and sales preparation.
This thesis uses the 2007.10~2009.3 insurance transactions of some insurance company in Taiwan to extract the humanities variables and policy variables, which are combined with the Executive Yuan Council for Economic Planning and Development of the economic signals variables. Then we utilize the decision tree technology in data mining to build forecast models for five different types of investment insurance products. The models can be used to adjust the sales strategies of investment insurance products in the future in accordance with the economic changes in the target-off sales of goods and the reference groups. The experimental results reveal that the accuracies of the prediction models for the five investment insurance products are all more than 85%. The modesl can be used by decision-makers of insurance companies for making the sale strategies of target customers in the future.
英文部分
[1] Berry ,M. J. A. and Linoff,G. S. Data mining techniques: for marketing, sales, andcustomer support, New York: Wiley Computer, 1997.
[2] Fayyad, U., Piatetsky-Shapiro, G. & Padhraic, S. From data mining to knowledgediscovery in databases. AI Magazine, 37-54,1996
[3] Frawley ,W. J., Paitetsky-Shapiro, and Matheus,G.C.J., “Knowledge Discovery inDatabase: An Overview”, California, edited by Piatetsky-Shapiro.G and Frawley ,W. J,AAAI/MIT Press, pp.1-30, 1991.
[4] Greenfeld, N. “Data Mining,” UNIX Review, pp.9-14, 1996.
[5] Grupe,F.H and Owrang,M.M., “Database Mining Discovering New Knowledge and Cooperative Advantage,” Information Systems Management, pp.26-31, 1995.
[6] Han, J., and Kamber, M. “Data mining:Concepts and techniques,”San Francisco:Morgan Kaufmann Publishes, pp.5-9, 2001.
[7] Peacock, P.R.“Data Mining in Marketing:Part I,” Marketing Management, Vol.6, No. 4, pp.818, 1998.
[8] Pyle, D. (1999). Data preparation for data mining, Morgan Kaufmann Publishers.
中文部分
[9] 尹相志 ,Microsoft SQL Server 2005 Data Mining 資料採礦與Office 2007 資料採礦增益集,悅知文化,台北市,2007 。
[10] 鍾碧玉 ,「投資型保險商品購買預測之研究」,國立台北大學企業管理學系碩士論文,2006
[11] 林哲鋒 ,「影響消費者購買投資型保險決策因素之研究-以台中縣市為例」,朝陽科技大學保險金融管理系碩士論文,2005
[12] 周紹康 ,「投資型保險購買意願因素之研究」,銘傳大學財務金融學系士論文,2004。
[16] 張家鳳 ,「利用資料挖掘技術建構保險業之差異化行銷模型」,世新大學資訊管理學系碩士論文,2004。
[17] 彭文正譯,Berry and Linoff 著,資料採礦:顧客關係管理暨電子行銷之應用,維科出版社,台北市, 2001。
[18] 曾憲雄、蔡秀滿、蘇東興、曾秋蓉、王慶堯著,資料探勘 Data Mining,旗標出版股份有限公司,台北市,2006 。
[19] 謝耀龍,壽險行銷,第三版,台北: 華泰文化,2004。
[20] 保險事業發展中心,投資型保險商品業務員訓練教材,2007
[21] 保險事業發展中心網址,http://www.tii.org.tw
[22] 行政院經濟建設委員會景氣指標查詢系統, http://index.cepd.gov.tw/
[23] 行政院經濟建設委員會網站http://www.cepd.gov.tw/
[24] 簡彬任,「消費者購買投資型保險商品之評估準則關鍵因素─層級分析法應用」,東海大學/企業管理學系碩士論文,2008。
[25] 吳鴻麟,「投資型保險在傳統通路與銀行通路行銷策略之研究--以A人壽保險公司為例」,國立台北大學/國際財務金融系碩士論文,2005。