Application of variable selection and dimension reduction on predictors of MSE’s development

Journal of Big Data

Table 11 Logistic regression for development status of enterprise based on principal factors

Full model					Reduced model by stepwise elimination
Coefficients	Estimate	Std. error	z value	Pr(\(>\|\hbox {z}\|\))	Coefficients	Estimate	Std. error	z value	Pr(\(>\|\hbox {z}\|\))
(Intercept)	− 2.897	0.933	− 3.107	0.002	(Intercept)	− 2.365	0.440	− 5.380	0.000
x1	0.017	0.016	1.023	0.306	x1	Removed
x2	− 0.092	0.044	− 2.099	0.036	x2	− 0.110	0.041	− 2.680	0.007
x3	− 0.069	0.070	− 0.981	0.327	x3	− 0.076	0.042	− 1.822	0.068
x4	− 0.026	0.050	− 0.514	0.608	x4	Removed
x5	− 0.091	0.114	− 0.799	0.424	x5	Removed
x6	− 0.161	0.119	− 1.351	0.177	x6	Removed
x7	0.138	0.129	1.070	0.285	x7	Removed
x8	− 0.217	0.122	− 1.783	0.075	x8	− 0.201	0.080	− 2.502	0.012
x9	0.138	0.051	2.725	0.006	x9	0.090	0.029	3.092	0.002
x10	− 0.151	0.139	− 1.087	0.277	x10	− 0.258	0.095	− 2.724	0.006
x11	0.448	0.176	2.541	0.011	x11	0.523	0.125	4.171	0.000
AIC	134.680				AIC	128.340