Monitoring and Analysis of Venture Capital and Corporate Fraud Based on Deep Learning

Table 9

Regression result based on PSM.

	(1)	(2)	(3)	(4)
	Radius-nearest neighbor (1 : 1) matching		Radius-nearest neighbor (1 : 2) matching
	VC	−0.084	−0.152	−0.076	−0.159
	(−1.92)	(−2.82)	(−2.06)	(−3.52)
Age	−0.058	−0.099	−0.057	−0.085
	(−2.00)	(−2.76)	(−2.42)	(−2.88)
Size	0.025	0.107	0.015	0.084
	(1.01)	(3.55)	(0.77)	(3.43)
ROA	−3.562	−5.654	−3.566	−6.267
	(−6.95)	(−9.24)	(−8.35)	(−12.51)
Dual	−0.048	−0.128	−0.043	−0.110
	(−0.97)	(−2.17)	(−1.06)	(−2.25)
Audit	−0.305	−0.335	−0.214	−0.234
	(−2.55)	(−2.19)	(−2.22)	(−1.90)
BdSize	0.222	0.196	0.230	0.198
	(1.84)	(1.34)	(2.35)	(1.67)
Tqmed	−0.007	−0.023	−0.040	−0.075
	(−0.07)	(−0.18)	(−0.50)	(−0.75)
Cash	−0.081	−0.139	−0.067	−0.120
	(−2.91)	(−4.01)	(−2.88)	(−4.19)
Constant	−1.055	−2.501	−0.758	−1.748
	(−1.56)	(−3.15)	(−1.41)	(−2.79)
Observations	4962	5002	7191	7226
Year&Industry	Yes	Yes	Yes	Yes
Pseudo R^2	0.046	0.054	0.039	0.050
AIC	4919.764	7447.950	7103.497	10869.554
Log likelihood	−2373.882	−3629.975	−3463.748	−5339.777

p < 0.10, p < 0.05, and p < 0.01.