Skip to main content

Table 3 Zero-inflated negative binomial model: robustness checks

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

(1) (2) (3) (4) (5) (6)
Control group   Adults aged 18 to 26 Childless adults Adults without the elderly Employed adults up to 65 years
   Number of obs = 494 Number of obs = 1243 Number of obs = 1471 Number of obs = 1117
   Non-zero obs = 130 Non-zero obs = 735 Non-zero obs = 837 Non-zero obs = 579
   Zero obs = 364 Zero obs = 508 Zero obs = 634 Zero obs = 538
   LR χ 2(6) = 3.60 LR χ 2(6) = 103.38 LR χ 2(6) = 38.56 LR χ 2(6) = 20.64
   Prob >χ 2 = 0.7301 Prob >χ 2 = 0.0000 Prob >χ 2 = 0.0000 Prob >χ 2 = 0.0021
Count process      
  reform –0.249 –0.015 0.034 0.109
  dummy_child 0.172 –0.023 –0.072 0.133
  interaction -0.659 –0.246 –0.583 –0.851
  female –0.116 0.144** 0.368*** 0.355***
  members –0.029 –0.265*** –0.059* 0.016
  log_income_memb –0.158 –0.461*** –0.202*** –0.003
  _cons 3.142 7.608*** 3.675*** 0.872
Logit      
  reform 0.725 0.807*** 1.617*** 1.605***
  dummy_child 3.284*** 3.415*** 4.765*** 5.042***
  interaction –0.233 –0.106 –0.988 –1.020
  female –0.967*** –0.881*** –1.334*** –1.462***
  members 0.414** 0.660*** 0.471*** 0.333**
  log_income_memb 0.299 0.246 0.115 0.122
  _cons –5.532 –5.895** –4.809 –4.567
  ln α –0.443 –0.757 –0.219 –0.242
  α 0.642 0.469 0.804 0.785
  1. Notes: * p<=0.1, ** p<=0.05, *** p<=0.01
  2. In order to interpret coefficients of the log–linear model, exponentiating is necessary; exp(coeff)