本研究採用向量誤差修正模型探討美國標普500指數與貨幣供給M2、美國十年期與三個月期公債殖利率利差、美國投資級公司債信用利差、零售銷售數據以及每月非農就業新增人數等五個經濟變數間的關聯性,研究的資料期間為2000年1月至2018年12月之月資料。 實證結果顯示(1) 變數間有五組共整合關係,標普500指數與貨幣供給、零售銷售數據與非農就業數據呈正相關;而與美公債殖利率利差呈負相關。(2) 美公司債信用利差、零售銷售數據與貨幣供給皆「Granger領先」標普500指數。(3) 當向量誤差修正模型分別受到來自標普500指數本身、零售銷售數據以及非農就業數據的衝擊時,標普500指數會呈現正向反應;當模型受到來自美公司債信用利差與美公債殖利率利差的衝擊時,標普500指數則會呈現負向反應。(4)預測誤差變異分解的結果為總體經濟變數對標普500指數價格波動的解釋力在第36個月時達47%,顯示對美國市場而言,總體經濟變數的衝擊會對美股價格造成一定程度的影響。其中又以標普500指數本身、零售銷售數據以及美公司債信用利差的解釋力最佳。
This study applies the vector error correction model (VECM) to investigate the relationship between the US S&P 500 Index and five macroeconomic variables (money supply M2, the spread between 3-month and 10-year treasury yield, US investment-grade credit spread, retail sales and nonfarm payroll) over the period from January 2000 to December 2018. The empirical results demonstrate that (1) there are five cointegration relationships. The S&P 500 Index is positively related to money supply, retail sales and nonfarm payroll, and is negatively related to the US treasury yield spread. (2) US investment-grade credit spread, retail sales and money supply do “Granger-cause” the S&P 500 Index. (3) When the VECM model faces the shock from the S&P 500 Index, retail sales and nonfarm payroll, the S&P 500 Index will react positively. However, when the model encounters the shock from the US investment-grade credit spread and the US treasury yield spread, the S&P 500 Index will react negatively. (4) The result of the forecast error variance decomposition shows that the macroeconomic variables’ explanatory power toward S&P 500 Index reaches 47% after 36 months. This indicates that for the US market, the shock from the macroeconomic variables influences the US stock prices to a certain degree. Moreover, the S&P 500 Index, retail sales and US investment-grade credit spread have better explanatory power among all the other variables.