Announcement

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

  • Kernel propensity score matching d-i-d technique in an unbalanced panel with different treatment years for each id


    Hi,

    I am working on a firm-level panel data (unbalanced panel). I want to estimate the impact of entering into import market on total factor productivity of firms using the user-written command "diff" in stata. My treatment group contains all those firms which enter the import market for the first time in a particular year and remain an importer for a minimum of two years. Also, the firms should have an observed history of atleast two years in the data prior to import entry. To get a value for the counter-factual, I am considering a control group that consists of firms which never entered the import market. However, I have a doubt in creating the variable "time"? Ideally, this variable takes a value 1 in the follow up period and 0 in the baseline (i.e. before treatment). I know how to substitute the value for the treated firms in this case, however, I am not able to figure out how should I assign the values for the controlled group. Please help me in clarifying this doubt. I shall be highly obliged.

    Thank You so much.
Working...
X