Can news-based economic sentiment predict bubbles in precious metal markets?

Financial Innovation

Table 4 Zivot-Andrews (ZA) stationary test

	Including intercept		Including intercept and trend
	Test statistics	Break date	Test statistics	Break date
NESI	− 4.6953**	2007:M06	− 4.7512**	2007:M08
NESI	(0.0342)		(0.0237)
MCSI	− 4.3398***	2007:M08	− 4.8534***	2007:M08
MCSI	(0.0052)		(0.0025)
SIBW	− 4.4101***	2006:M10	− 4.6011***	2006:M10
SIBW	(0.0008)		(0.0000)
Inflation	− 4.8220***	2008:M10	− 4.8519***	2008:M09
Inflation	(0.0089)		(0.0008)
USDI	− 4.9993***	2008:M02	− 4.9500***	2007:M08
USDI	(0.0001)		(0.0001)
EFR	− 3.4722**	2008:M02	− 3.7061***	2008:M10
EFR	(0.0166)		(0.0040)
T-Spread	− 3.1510***	2007:M08	− 3.5337***	2007:M08
T-Spread	(0.0087)		(0.0037)
GEA	− 4.8605***	2007:M06	− 4.8512***	2007:M07
GEA	(0.0004)		(0.0018)

The table reports the Zivot-Andrews (ZA) statistics, which allows for both a structural break in intercept, trend or both. The null hypothesis of the ZA test is that the series has a unit root with a structural break(s) against the alternative hypothesis that they are stationary with a break(s). p values are given in brackets. *, **, and *** indicate significance at 10%, 5%, and 1% levels, respectively