Skip to main content

Table 13 Probit results with additional control variables

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

 

Dependent variable: Bubble

Gold

Silver

Palladium

Platinum

\(NESI_{t - 1}\)

0.0702**

0.0036

0.0206

0.0474***

(0.0154)

(0.7690)

(0.7400)

(0.0000)

\(Inflation_{t - 1}\)

1.6046*

0.6159*

1.3735***

1.8759***

(0.0940)

(0.0977)

(0.0000)

(0.0080)

\(US{\text{DI}}_{t - 1}\)

− 1.6744**

− 1.6348***

− 2.8399***

− 1.8218***

(0.0010)

(0.0000)

(0.0000)

(0.0049)

\({\text{EFR}}_{t - 1}\)

− 0.0031*

− 0.2431*

− 0.5383***

− 0.1161***

(0.0970)

(0.0890)

(0.0000)

(0.0020)

\(T - Spread_{t - 1}\)

− 0.0479*

− 0.4905*

− 0.2297*

− 0.3733**

(0.0887)

(0.0680)

(0.0519)

(0.0235)

\({\text{GEA}}_{t - 1}\)

0.0083***

0.0070**

0.0046***

0.0049***

(0.0000)

(0.0260)

(0.0032)

(0.0030)

\({\text{EPU}}_{t - 1}\)

0.0128**

0.0037

0.0045**

0.0264***

(0.0130)

(0.6130)

(0.0362)

(0.0000)

\({\text{GPR}}_{t - 1}\)

0.0086***

0.0014

0.0120***

0.0006**

(0.0070)

(0.7910)

(0.0000)

(0.0409)

\({\text{ID}} - {\text{EMV}}_{t - 1}\)

0.0254***

− 0.0094

0.0012

0.0040

(0.0000)

(0.7280)

(0.9150)

(0.6020)

\(Constant\)

0.6003

0.1959**

0.7914***

− 0.1788*

(0.1560)

(0.0120)

(0.0000)

(0.0554)

\(Observations\)

408

408

408

408

McFadden's pseud-R2

0.7546

0.4845

0.4657

0.6786

Log-likelihood

− 88.0272

− 67.0566

− 144.0302

− 85.3862

Hosmer–Lemeshow test

6.030

1.570

2.250

5.960

(0.4193)

(0.6670)

(0.1332)

(0.1508)

  1. The dependent variable is a binary that equals 1 (bubble dates) and 0 (none-bubble dates) identified by the GSADF procedure. The Hosmer–Lemeshow test is a statistical test for goodness of fit for probit regressions, following the \(\chi^{2}\) distribution. A large \(\chi^{2}\) value (with small p value \(< 0.05\)) indicates poor fit regression model. p values are given in brackets.*, **, and *** indicate significance at 10%, 5%, and 1% levels, respectively.