Industrial Demand and Innovation: An Application of Binomial Regression Model to Project Statistics of NSFC of China

Table 4

Regression estimation.

Explained variable: total number of approved projects
	(1) OLS	(2) OLS	(3) OLS	(4) NBR	(5) NBR	(6) NBR
ln (Pa)		0.002	0.003		0.505	0.227
		(0.001)	(0.002)		(0.046)	(0.120)
ln (Ph + Pd)		0.004	0.014		0.889	0.603
		(0.003)	(0.003)		(0.114)	(0.849)
	0.002		−0.006	0.621		0.737
	(0.0003)		(0.007)	(0.126)		(0.368)
ln (Pm)	0.973		0.002	0.758		0.747
	(0.093)		(0.001)	(0.441)		(0.107)
ln (Ar)			0.077			0.204
			(0.018)			(0.047)
Constant term	−1.593	−0.790	−0.054	1.951	2.256	2.046
	(0.831)	(0.852)	(0.876)	(0.044)	(0.039)	(0.048)
N	483	483	483	483	483	483
R^2	0.916	0.894	0.929
Log likelihood				−1658.582	1675.872	−1625.618

Data source: National Natural Science Foundation of China.