###### Do File STATA – Probit Model
March 3, 2014
March 5, 2014
• x1, x2 and x3 are more likely to have a contract with middlemen, and x4, x5 and x6 are less likely to have a contract.
• Marginal effect (ME): x1, x2 and x3 are N% more likely to have a contract with middlemen.
• The predicted probabilities are limited between 0 and 1.
• Probit are estimated using the ML method.
• an increase in x increases/decreases the likelihood that y=1
• an increase in x makes the outcome of 1 more likely.
• We interpret the sign of the coefficient but not the magnitude.
• The magnitude can’t be interpreted using the coefficient because different model has a different scales of coefficient.
• Interpret ME: an increase in x increase the probability y=1, by the marginal effect expressed as a percent. We can interpret both sign and magnitude.
• Predicted probability indicate the likelihood of y=1, if the predicted prob is greater than 0.5 we can predict that y=1, others y=0. ##### Dias Satria
Dias Satria, Research field :Economic development, international trade, Banking and small/medium enterprise Email. dias.satria@gmail.com Mobile Phone. +62 81 333 828 319 Office Phone. +62 341 551 396