Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Formally writing an IV regression

    Hello, I am new to Statalist and am currently writing my dissertation. I’m sorry if any of this question doesn’t fit the statalsit guidelines, just let me know and I will change it next time I post.

    My question is about how to write my Stata command as a formal formula in my document.

    If I have regression: regress Y X1 i.X2 X3 [pweight=weight]
    Where X1 is continuous, X2 is a categorical variable with 3 categories and X3 is a dummy variable then I believe I would write is as:
    Click image for larger version

Name:	formula.png
Views:	1
Size:	2.3 KB
ID:	1438131




    In my case, I have an IV regression:

    ivreg2 Y X2 V (X1 X1#X2 = i.Z1) [pweight=weight]
    where Y is my dependent variable, X1 is my key explanatory variable which is endogenous due to reverse causality between it and Y and is a dummy variable, X2 is my treatment effect and is also a dummy variable, Z1 is an instrument for the endogeneity and V is a vector of control variables.

    How do I formally transform this Stata command into a formula which I can write in my dissertation as I have done in the simple regression example? My thoughts were that I might write it as 2 formulas. 1. a normal regression 2. A regression showing the endogeneity issue but then I haven’t included Z1 in the regressions…

    To save me explaining my data and to make the question relevant to all readers I won’t go into the specifics of my data set but if you want any further info to answer my question, please just ask.

    Thank you so much!!
    Last edited by Lucy Ty; 07 Apr 2018, 08:48.
Working...
X