Financial Innovation

Table 2 Probit robustness checks estimating pre-chasm usage of robo-advice

From: Overconfidence and the adoption of robo-advice: why overconfident investors drive the expansion of automated financial advice

Dependent	(1)	(2)
Variable: robo-advice	Robustness OC probit	Parker&Stone probit
Robustness overconfidence	0.42***
	(0.06)
Confidence		0.35***
		(0.05)
IRT knowledge		− 0.39***
		(0.06)
Willingness to take risk	0.14***	0.11***
	(0.03)	(0.03)
Female	0.01	− 0.08
	(0.09)	(0.09)
Age categories
25–34	− 0.08	− 0.01
	(0.19)	(0.20)
35–44	− 0.37	− 0.23
	(0.20)	(0.21)
45–54	− 0.82***	− 0.60**
	(0.20)	(0.21)
55–64	− 0.87***	− 0.55*
	(0.20)	(0.22)
65+	− 1.31***	− 0.97***
	(0.22)	(0.23)
Bachelor or more	− 0.06	0.05
	(0.10)	(0.10)
Household income
$50,000–$100,000	− 0.18	− 0.11
	(0.12)	(0.12)
$100,000 or more	− 0.52***	− 0.37**
	(0.14)	(0.14)
Married	0.20*	0.11
	(0.10)	(0.10)
Ethnic/racial minority	0.18	0.10
	(0.10)	(0.10)
Constant	− 1.48***	− 3.31***
	(0.28)	(0.35)
Observations	1,923	1,937
McFadden's pseudo R²	0.275	0.318
BIC	1,199	1,145
AIC	1,121	1,061

Column (2) illustrates separate effects of IRT knowledge and confidence in one’s investment knowledge (Parker and Stone 2014). IRT Knowledge (2) are Bayesian mean values of the latent trait. Robust standard errors in parentheses
***p < 0.001; **p < 0.01; *p < 0.05

Back to article page