道瓊運輸指數、工業指數與總體經濟變數關係之研究

The Dow Jones Transportation Index, Industrial Index and Macroeconomic Variables-An Empirical Evidence

楊俐玲(Li-Ling Yang)

指導教授 ： 謝德宗

國立臺灣大學/社會科學院/經濟學研究所/碩士(2016年)

https://doi.org/10.6342/NTU201603649

摘要

本文係以道瓊運輸指數、道瓊工業指數及總體變數為研究對象，探討道瓊運輸指數和美國股價及總體變數的關聯性，選取的總體變數包括PMI製造業指數、新屋開工數字、零售銷售數字及失業率。
研究架構係採樣自2000年01月至2015年10月之月樣本資料，以單根檢定、向量自我迴歸模型(VAR)、Granger因果檢定、衝擊反應函數及預測誤差變異數進行實證分析，首先以單根檢定確認各變數取對數並經一階差分後為定態數列後，建立變數間向量自我迴歸模型觀察道瓊運輸指數與股價及總體變數落後期之相關性，由衝擊反應函數分別了解外生衝擊對道瓊運輸指數與股價的影響，並以Granger因果關係檢定變數間是否存在領先、落後、互相領先，或兩者無任何關係，最後以預測變異數誤差分解了解各變數之間的解釋能力。研究結果顯示:道瓊運輸指數對道瓊工業指數無解釋能力，道瓊工業指數對道瓊運輸指數解釋能力高，可解釋此二指數走勢同向且一致，道瓊運輸指數和道瓊工業指數並無「領先-落後」關係，有運輸需求的新屋開工數、PMI製造業指數及零售銷售數和道瓊運輸指數關聯性大，且領先道瓊運輸指數。

關鍵字

向量自我迴歸模型 ； Granger因果關係檢定 ； 衝擊反應函數 ； 預測變異數誤差分解

並列摘要

This paper discusses the relationship between the Dow Jones transportation index、 the Dow Jones industrial average index and the macroeconomic variables, including retail sales, purchasing managers ' index, new house starts and unemployment rate in US. We use quantitative methods such as Unit Root Test,Vector Autoregression Model,Granger Causality Test, Impulse Response Analysis and Forecast error Variance Decomposition. This research collected monthly data ranging from January 2000 to October 2015. First, we use unit root test to ensure all the series used in regression analysis are stationary. Then, we create the vector autoregression(VAR) model to analyze how explanatory variables affect that the Dow Jones industrial average index and The Dow Jones transportation index ,and use impulse response function to figure out the response of the Dow Jones transportation index and the Dow Jones transportation index to the exogenous shock of another variables. Granger causality test is also used to determine whether a time series is useful in forecasting another. Eventually, use the Forecast Error Variance Decomposition to indicates that how much of the forecast error variance of each of the variables can be explained by exogenous shocks to the other variables. The conclusion of this research shows that the Dow Jones transportation index movements explain a rarely fraction of the forecast error variance in the Dow Jones Industrial Average Index. The Dow Jones Industrial Average Index movements explain a larger fraction of the forecast error variance in the Dow Jones transportation index than itself. This result can explain the trend of the two index are consistent. The Dow Jones transportation index Granger cause the Dow Jones Industrial Average Index is not significant. The variables with transportation demands, the new house starts, the retail sales and the PMI , are Granger cause and play important roles in affecting the Dow Jones transportation index.

並列關鍵字

Vector Autoregression Model ； Granger Causality Test ； Impulse Response Analysis ； Forecast error Variance Decomposition

參考文獻

8. 蕭堯仁(2011)，「波羅的海乾散貨運價指數與金磚四國股價之關聯性」，航運季刊，第20卷第4期，第21-24頁。

12. Chen , Nai–Fu ,Richard Roll , Stephen A.Ross(1986), “Economic Forces and the Stock Market” ,The Journal of Business, p383-p403.

7. 劉祥熹、涂登才(2012)，「美國股市及其總體經濟變數間關連性與波動性之研究」，經濟研究，48:1 (2012)，頁139-189。

10. Ang, G Bekaert, M Wei,(2007) ”Do macro variables, asset market , or surveys forecast inflation better? “ Journal of Monetary Economics 54(4),1163-1212.

13. Chen, Shiu-Sheng (2009) “Predicting the Bear Stock Market: Macroeconomic Variables as Leading Indicators,” Journal of Banking and Finance, 33:2, 211-223.