Table A3

IVprobit regressions: Islamic religiosity proxies and TEA (omitted variable bias)

VariablesModel 8Model 9Model 10Model 11Model 12
Age−0.000239−0.000183−0.000227−0.000282−0.000405
(0.000915)(0.000914)(0.000916)(0.000919)(0.000938)
Education0.0460***0.0472***0.0454***0.0462***0.0442***
(0.00949)(0.00949)(0.00950)(0.00952)(0.00983)
Gender0.001450.002260.001830.001620.00731
(0.0206)(0.0206)(0.0207)(0.0207)(0.0211)
Internet0.0157***0.0186***0.0167***0.0165***0.0214***
(0.00155)(0.00173)(0.00160)(0.00159)(0.00194)
Construction_Worker−1.91e−06−1.61e−06−1.01e−06−1.54e−06−3.57e−06**
(1.23e−06)(1.20e−06)(1.15e−06)(1.24e−06)(1.58e−06)
Ln_Investment0.0462***0.0430***0.0537***0.0630***0.0362***
(0.0123)(0.0128)(0.0117)(0.0108)(0.0132)
NPL−10.46***−13.22***−12.49***−12.30***−12.94***
(1.512)(1.641)(1.596)(1.600)(1.571)
NPF−0.8870.421−0.0425−0.5222.103***
(0.595)(0.630)(0.607)(0.597)(0.646)
Population_density−2.89e−05***−2.94e−05***−3.40e−05***−2.79e−05***−2.79e−05***
(4.62e−06)(4.63e−06)(4.93e−06)(4.62e−06)(4.97e−06)
Ln_GDP0.188***0.243***0.179***0.162***0.172***
(0.0377)(0.0451)(0.0374)(0.0360)(0.0417)
Ln_Islamic_school0.0967***    
(0.0192)    
Ln_Islamic_seminary 0.0757***   
 (0.0154)   
Ln_Hajj  0.0850***  
  (0.0179)  
Ln_Mosque   0.0690*** 
   (0.0148) 
Ln Halal    0.128***
    (0.0277)
Constant−4.335***−4.754***−4.430***−4.061***−4.428***
(0.454)(0.521)(0.481)(0.436)(0.550)
Observations21,27921,27921,27921,27920,440
CLR (p-value)0.00000.00000.00000.00000.0000
AR (p-value)0.00000.00000.00000.00000.0000
LM (p-value)0.00000.00000.00000.00000.0000
LM-J (p-value)<0.05<0.05<0.05<0.05<0.05
Sargan (p-value)0.0180.00990.00570.00210.878

Note(s): Standard errors in parentheses

***p < 0.01, **p < 0.05, *p < 0.1

Dependent variable: a dummy variable, either a respondent is involved or not in early-stage entrepreneurial activity (0 = No, 1 = Yes). Independent variables: The current age of the respondents (in years) (Age), UN harmonized educational attainment (Education), The gender of the respondent (0 = male, 1 = female) (Gender), The percentage people with age more than 10 years who access internet in the last 3 month based on province within service industry (Internet), Skilled construction worker at province level (Construction worker), Natural logarithm foreign direct investment realization by Province (million US$) (Ln Investment), Percentage non-performing loan to total amount of loan in large conventional banks at province level (NPL), Percentage non-performing financing to total amount of financing in large Islamic banks at province level (NPF), Population density per sq.km by province (Population density), Natural logarithm per capita gross regional domestic product at current market prices by Province (thousand rupiahs) (Ln GDP), Natural logarithm sum of Islamic primary school, Islamic junior high school, and Islamic senior high school at the provincial level (Ln Islamic school), Natural logarithm of the number of Islamic seminary school (pesantren) by province (Ln Islamic seminary), Natural logarithm of the number of Hajj application by province (Ln Hajj), Natural logarithm of the number of Mosque registered under Ministry of Religious Affair at the provincial level (Ln Mosque), Natural logarithm of the number of halal certification at the provincial level (Ln Halal)

Models 8, 9, 10, 11, and 12 are estimated using IVprobit regression. Instrumental variables: (1) lagged 1 year natural logarithm of the number of natural disasters by province, and (2) lagged 1 year natural logarithm of the number of houses destroyed/damaged by natural disaster

The p−values for the CLR and AR are against the null hypothesis that the coefficient of the endogenous variable, i.e. lagged one year natural logarithm of natural disaster events and lagged one year natural logarithm of houses being destroyed and damaged by natural disasters, is zero

Source(s): Authors’ own work

or Create an Account

Close Modal
Close Modal