本文選用Bloomberg資料庫中的債券資料庫以及美國勞工統計局BLS公佈之各項就業指標,並使用各項時間序列計量方法,如單根檢定、Granger因果關係簡定、向量自回歸模型及衝擊反應函數等等,探討(1)各項就業數據與債券殖利率間是否存在領先或落後之關係、(2)各項就業數據相對債券殖利率而言為正或負向關係。 實證結果顯示僅JOLT調查職位空缺數月變動可Granger影響債券殖利率利差,但非農就業月變動及失業率月變化亦影響JOLT職位空缺數月變動。另一方面,失業率月變化與JOLT調查職位空缺數月變動與債券殖利率利差顯著相關,但失業率月變化與債券殖利率利差同向變化、JOLT調查職位空缺數月變動與債券殖利率利差反向變化。
In this article, we use the bond yield data from the Bloomberg database and various employment indicators published by BLS (Bureau of Labor Statistics). Using multiple time-series measurement methods, such as unit root test, Granger causality test, vector autoregressive model and impulse response function, to prove our idea. First, whether there is a leading or lagging relationship between each employment indicators and bond yields. Second, whether each employment indicators and bond yields are positively or negatively related. The empirical results show that monthly change of JOLTs output is Granger cause the spread of 10-year bond yields, and monthly change of NFP and unemployment rate are Granger cause the monthly change of JOLTs output. On the other hand, monthly change of unemployment and JOLTs output are significantly related to the spread of 10-year bond yields. Monthly change of unemployment rate is positively related, but the other is negative related to spread of 10-year bond yields.