Dependant / Independant / Instrumental / Control variable

[Abdullah Baba]

Hi, can anyone help me with the following probit regression and clarify which one stands for which variable?

probit ceo_suc Wni_asset_trend i2a_bhy_l_trend ln_at_trend i_boardsize i_pid pred_ceo_duality founder cum_no_se yd_* id_* if sample==1, vce(cluster firm_id)

Thank you

[Amin Sofla]

I think that you should provide more information in order to enable us to answer your question properly. The below are my guess:
The probit command runs the Probit regression, also called a probit model. This model is used to model binary outcome variables.
Probit model with endogenous regressors can be tested using ivprobit, etc. For the instrumental variable using probit, you might also want to take a look at notes 1 and 2.
This dataset has a binary response (outcome, dependent) variable called ceo_suc.
Wni_asset_trend , i2a_bhy_l_trend , ln_at_trend , i_boardsize , i_pid , pred_ceo_duality , founder , cum_no_se are among your independent variables. With the provided information, I cannot find out which independent variables are your main independents. You should check the article's main reseach question.
yd_* most likely are the dummy variables for years.
id_* most likely are the dummy variables for individuals.
if sample==1 limits the analysis to the sample that equals to 1.
vce(cluster clustvar) specifies that the standard errors allow for intragroup correlation, relaxing the requirement that the observations be independent. Here, your cluster variable is firm_id.

[Phil Bromiley]

To add to Amin's comment, you'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.
To answer your specific question, in this model ceo_suc is the dependent variable and all the other variables are assumed to be independent (exogenous) variables. Amin gave you guidance on where to look if you have endogeneity. Any variable for which you have not offered a hypothesis is often considered a control.

[Abdullah Baba]

Hi Amin Sofla,

thank you so much for the very detailed explanation. This helps me much more than I actually expected. Thanks again!!

Phil Bromiley Thank you too!.

Have a good evening!