Probit model jika ada masalah variabel yang endogen?

Bagaimana membuat tabel di STATA
May 11, 2015
E-Money (Poster)
May 14, 2015
Show all

Bagaimana jika salah satu variabel independents adalah endogen?

Step:
1) Identify which of you vars are endogenous (see which ones that you use in the SEM model as dependent vars).
2) Find instruments. Lit review would help. eg distance vars (distance to market, distance to village leader) are often used as instruments for cost and other institutional/governance vars.
3) run ivprobit as defined below.
=============================================
2) Endogeneity
One may argue that there is a two way relationship between these variables and the presence of contract. Mathematically:
Z*=x’B+u
Z* is latent variable; Z=1 (observable) when contract is present and u is the error term
Endogeneity: Corr(x,u)!=0 or correlation between x and u is not zero.

I would suggest to see the literature how they specify this contract participation model and identify which ‘x’ that is/are endogenous. For example, the theory suggests monitoring costs is endogenous, meaning it can influence one’s decision on participating at contract but the contract participation can affect his/her monitoring costs.
If that is the case, you can use ‘ivprobit’
1st stage: determinants of monitoring costs (OLS)
2nd stage: probit model of contract
http://www.stata.com/manuals13/rivprobit.pdf

Dias Satria
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

Leave a Reply

Your email address will not be published. Required fields are marked *