Announcement

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

  • static staggered TWFE difference in difference model

    I am trying to use a static staggered TWFE difference in difference model. I have house price transaction data for all house sales in two counties in the UK since 2003 to September 2022. 22 parishes within these counties implemented a policy which bans the sale of new build second homes and I am trying to assess the effect this has on house prices. However, the parishes all implemented the policy at different times. My plan is to tun this regression:

    PricePti= βPostPt+ γP+ λt+ εPti

    Where Post is a dummy variable which captures whether parish P has implemented the policy by year t.

    γP is the unit fixed affect at the parish level and λt is the time fixed effect at the monthly level because the policy was introduced in particular months of different years in different parishes.
    My question is what command should I use to run this regression in stata and should I use the log of Price because that gives me a nicer normal distribution of the variable?

    I was thinking would i use the didregress command:
    didregress (Price) (post), group(iCode) time(transaction_month)
    Where the variable iCode lists a code for each transaction which corresponds to which parish it is in and transaction_month is a variable which gives the month and year of the transactions.

Working...
X