臺灣類股指數預測與輪動之實證研究

An Empirical Study of Taiwan Stock Index Forecast and Rotation

陳芝吟(CHEN, CHIH-YIN)

指導教授 ： 古永嘉

國立臺北大學/商學院/企業管理學系/碩士(2010年)

摘要

資訊科技發達，投資人能輕易獲得各種的分析方法與資訊，但仍要面臨最佳投資進出時點作決定，與如何取捨何種投資建議的困擾。然而，股票證券市場素有經濟櫥窗之稱，如果没有藉由景氣指標來掌握整體景氣循環脈動，選擇對的股票在對的時間點進行投資或處分，即使是選對產業、選對標的、做好最適配置資產組合，仍無法獲得預期的報酬，可見掌握投資標的的趨勢與時機點之重要性。
本研究以臺灣景氣指標與類股指數為研究對象，資料來源取自『行政院經濟建設委員會』及『臺灣證券交易所』，研究期間為1999年1月至2009年12月，共十一年月資料，132個樣本。本研究將市場股價指數、電子工業類股指數、電機機械類股指數、電器電纜類股指數、金融保險類股指數、機電類股指數設為依變數，景氣指標設為自變數，利用逐步迴歸分析與古典時間序列分析-X11季節分解法進行實證分析。結果發現：
一、 落後一期至六期之景氣指標對於當期之市場及類股指數，具有顯著之影響性及預測性。
二、 市場及各類股指數具季節性波動與淡旺季特性。
三、 可藉由市場及各類股指數的長期趨勢及週期性，預測其輪動關係。

關鍵字

景氣指標、類股指數、逐步迴歸分析、古典時間序列分析-X11

並列摘要

Investors can easily obtain various analysis methods and financial advices in this well-developed information technology environment, but making the optimal investing decision can still be confusing with numerous investing suggestions on the market. However, stock markets provide a foreseeable financial development for investors to estimate the perfect timing to process the commerce as gateway to comprehend the macroeconomic. Somehow it is not enough to have the predicted remuneration even with all the best combination of investing principles, therefore it shows how important it is to monitor the investing trend and when is the timing at its peak.
This study sets the stock index as the final result, the business indicator as a cause, applying to The Stepwise Regression Analysis and The Classic Time Series Decomposition-X11, as the result, the variable behavior mutually influences each other. The material is originally taken from & quoted『Business indicators database, Council for Economic Planning and Development. R.O.C. Taiwan』and 『Taiwan stock exchange』, the study for the period of Year 1999 to 2009, a total of eleven years with 132 samples. The study will stock index rate-of-located in the best variable, set the quarter of a pointer to business indicators is set to variable. Empirical results found:
1. Early stage form one to six between a period of business indicators means the stock index of the current period against the forecast have significant impact.
2. The various stock index has the seasonal fluctuation and the off season or busy season.The various stock index's long-term trend and the periodicity, may utilize in forecast that wheel of the various stock moves relations.

並列關鍵字

Business indicators, Stock index, Stepwise Regression Analysis, Classical Time Series Decomposition-X11

參考文獻

6. 林大超（2003），台灣與美國兩地景氣循環指標關聯性之研究，國立成功大學管理學院未出版之碩士論文。

8. 林金葉（2009），電信業財務五力與股價報酬率關聯性之研究-以電信三雄為例，國立臺北大學企業管理學系未出版之碩士論文。

2. Bendel, R. B. & Afifi, A. A. (1977). Comparison of Atopping Rules in Forward Stepwise Regression, Journal of the American Statistical Association, 72, pp.46-53.

4. Duffee, G. R., & Prowse, S. (1996), “What’s Good for GM …? Usingauto Industry Stock Returns to Forecast Business Cycles and Test the Q -theory of Investment”, Finance and Economics Discussion Series papers.

5. Fama, E.F., (1970), Efficient Capital Markets : A Review of Theory and Empirical Work, Journal of Finance, 25(5), pp.383-417.