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

Wubetie, Habtamu Tilaye

doi:10.1186/s40537-018-0153-4

Journal of Big Data

Table 5 Linear regression for the number of employer in an enterprise based on selected factors through stepwise elimination and lasso methods

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

Variables selected by stepwise selection						Variables selected using lasso method						Reduced model
Coefficients	Estimate	Std. error	t value	\(\hbox {Pr}(>\|\hbox {t}\|)\)	Significance	Coefficients	Estimate	Std. error	t value	\(\hbox {Pr}(>\|\hbox {t}\|)\)	Significance	Coefficients	Estimate	Std. error	t value	\(\hbox {Pr}(>\|\hbox {t}\|)\)	Significance
(Intercept)	0.315	0.187	1.683	0.094	.	(Intercept)	0.231	0.163	1.421	0.157		(Intercept)	0.376	0.198	1.898	0.060	.
c3	− 0.298	0.150	− 1.985	0.049	*	h2	− 0.100	0.045	− 2.247	0.026	*	c3	− 0.280	0.152	− 1.843	0.067	.
c6	0.183	0.078	2.347	0.020	*	h4	0.292	0.083	3.522	0.001	***	h2	− 0.050	0.050	− 0.999	0.319
h3	− 0.311	0.080	− 3.900	0.000	***	h14	0.136	0.085	1.593	0.113		X30.49	1.951	0.042	46.677	0.000	***
h4	0.398	0.087	4.593	0.000	***	f2	− 0.190	0.094	− 2.023	0.045	*	emp_Female	2.126	0.060	35.406	0.000	***
f2	− 0.189	0.089	− 2.111	0.036	*	IF3	− 0.033	0.028	− 1.183	0.239		c6	0.190	0.078	2.425	0.016	*
IF1	− 0.035	0.022	− 1.580	0.116		IF6	− 0.027	0.020	− 1.378	0.170		h3	− 0.272	0.089	− 3.046	0.003	***
IF6	− 0.034	0.019	− 1.813	0.072	.	IF8	− 0.189	0.035	− 5.458	0.000	***	h4	0.403	0.087	4.626	0.000	***
IF8	− 0.219	0.033	− 6.583	0.000	***	Category	− 0.011	0.031	− 0.349	0.727		f2	− 0.188	0.091	− 2.073	0.040	*
s3	− 0.056	0.040	− 1.403	0.162		Grouping	0.266	0.113	2.359	0.020	*	IF1	− 0.034	0.022	− 1.555	0.122
s4	0.116	0.072	1.616	0.108		Emp_0	0.003	0.006	0.562	0.575		IF6	− 0.033	0.019	− 1.736	0.084	.
In1	− 0.067	0.038	− 1.778	0.077	.	X15.29	0.964	0.025	38.848	0.000	***	IF8	− 0.219	0.033	− 6.550	0.000	***
In2	0.128	0.071	1.806	0.073	.	X30.49	0.982	0.011	87.476	0.000	***	s3	− 0.056	0.040	− 1.400	0.163
In3	− 0.126	0.075	− 1.675	0.096	.	X50.65	0.486	0.113	4.305	0.000	***	s4	0.120	0.072	1.669	0.097	.
In7	0.118	0.074	1.601	0.111		X.65	0.313	0.201	1.555	0.122		In1	− 0.066	0.038	− 1.728	0.086	.
In10	− 0.100	0.073	− 1.373	0.172		ed0	− 0.057	0.106	− 0.539	0.591		In2	0.120	0.072	1.672	0.097	.
Grouping	0.384	0.090	4.276	0.000	***	ed2	0.027	0.026	1.047	0.297		In3	− 0.112	0.077	− 1.442	0.151
X15.29	1.007	0.020	51.029	0.000	***	emp_Male	0.950	0.047	20.421	0.000	***	In7	0.128	0.075	1.714	0.088	***
X50.65	1.543	0.098	15.827	0.000	***	emp_Female	1.071	0.048	22.330	0.000	***	In10	− 0.110	0.074	− 1.488	0.139
X.65	0.428	0.190	2.252	0.026	*							Grouping	0.390	0.090	4.322	0.000	***
ed0	− 1.034	0.096	− 10.763	0.000	***							X15.29	1.006	0.020	49.555	0.000	***
ed1	1.076	0.010	103.817	0.000	***							X50.65	− 0.590	0.105	− 5.625	0.000	***
ed3	− 0.058	0.030	− 1.917	0.057	.							X.65	0.378	0.197	1.922	0.056	.
emp_Male	1.953	0.041	47.527	0.000	***							ed0	0.920	0.066	13.968	0.000	***
												ed1	− 1.051	0.046	− 23.026	0.000	***
												ed3	− 0.053	0.032	− 1.633	0.104
AIC			− 280.540			Optimal lambda			0.052
Multiple R-squared			0.995			Multiple R-squared			0.994			Multiple R-squared			0.995
Adjusted R-squared			0.994			Adjusted R-squared			0.993			Adjusted R-squared			0.994
F-statistic			1231.000			F-statistic			1415.000			F-statistic			1125.000
P-value			\(<2.2{\rm E}{-}016\)			P-value			\(<2.2{\rm E}{-}016\)			P-value			\(<2.2{\rm E}{-}016\)
Significance codes: 0 ‘*’ 0.001 ‘’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1						Significance codes: 0 ‘*’ 0.001‘’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’1						Significance codes: 0 ‘*’ 0.001 ‘’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Back to article page