Skip to main content

Table 2 Zero-inflated negative binomial model

From: The abolition of user charges and the demand for ambulatory visits: evidence from the Czech Republic

 

Number of obs = 1841

 

Non-zero obs = 1176

 

Zero obs = 665

 

LR χ 2(6) = 79.38

 

Prob >χ 2 = 0.0000

 

Visits

Coef.

Count process = neg. binomial

  
 

reform

0.010

 

dummy_child

–0.242

 

interaction

–0.497

 

female

0.190***

 

members

–0.132***

 

log_income_memb

–0.337***

 

_cons

5.840***

Inflation model = logit

  
 

reform

1.232***

 

dummy_child

4.377***

 

interaction

–0.588

 

female

–1.233***

 

members

0.502***

 

log_income_memb

0.261

 

_cons

–6.292***

 

ln α

–0.451

 

α

0.637

  1. Likelihood-ratio test of α = 0: \(\bar {\chi }^{2}\)(01) = 1393.26 Pr\(\geq \bar {\chi }^{2}\) = 0.0000
  2. Vuong test of zinb vs. standard negative binomial: z = 6.91 Pr >z = 0.0000
  3. Notes: * p<=0.1, ** p<=0.05, *** p<=0.01
  4. In order to interpret coefficients of the log–linear model, exponentiating is necessary; exp(coeff)
Back to article page

Health Economics Review

ISSN: 2191-1991

Contact us