Table 6.

Estimated results for binary choice model using diabetes as dependent variable

Variable	OLS	Logit	Probit	Skewed Logit	Mixed effect Logit
Obesity dummy	0.065*	0.719*	0.307	0.697	0.719*
Marriage status	0.241*	1.660	0.978	1.434	1.660
Hypertension dummy	0.056*	0.764*	0.361	0.693	0.764*
Gender dummy	−0.038*	−0.670	−0.374	−0.607	−0.670
Ln(house building area)	−0.059**	−1.300**	−0.571**	−1.247**	−1.300**
Ln(income per month)	0.013	0.329	0.187	0.301	0.329
Constant	0.239	0.570	−0.336	−12.827	0.570
F /LR(Wald) test for all variables	3.06***	14.26**	13.5**		11.74**
R square/Pseudo R square	0.67	0.14	0.14
Adjusted R square	0.51
Log likelihood		−42.85	−43.23	−42.62	−42.85

Note: The null hypothesis for F or LR test is that the concerned coefficients are jointly equal to zero

*, **, and *** give the coefficients’ significance indicated by estimated standard errors at 10%, 5% and 1% level individually