This is my first time posting. I have read a number of super helpful posts (thank you Clyde Schechter) that have gotten me to this point but I am still in need of assistance and hoping for some guidance.
I am working with panel data from two cities where my key dependent variable is the MONTHLY violent crime rate so I have months starting in 2014 and into 2018. About mid way through these years one city received an intervention and the control city did not. I want to test whether the intervention (treat) had an impact on the monthly crime rate. Here is my model and results.
city2 is 1 for treated city and 0 for control city
id is the monthyear id
Code:
xtset city2 id xtreg Vio_R time##treat, fe cluster(city2)
Code:
. xtreg Vio_R time##treat, fe cluster(city2) note: 1.treat omitted because of collinearity Fixed-effects (within) regression Number of obs = 84 Group variable: city2 Number of groups = 2 R-sq: within = 0.4145 Obs per group: min = 42 between = 1.0000 avg = 42.0 overall = 0.2758 max = 42 F(0,1) = . corr(u_i, Xb) = -0.1550 Prob > F = . (Std. Err. adjusted for 2 clusters in city2) ------------------------------------------------------------------------------ | Robust Vio_R | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- 1.time | -10.51491 5.07e-14 -2.1e+14 0.000 -10.51491 -10.51491 1.treat | 0 (omitted) | time#treat | 1 1 | -5.391841 5.02e-14 -1.1e+14 0.000 -5.391841 -5.391841 | _cons | 32.47385 1.22e-14 2.7e+15 0.000 32.47385 32.47385 -------------+---------------------------------------------------------------- sigma_u | 7.5495077 sigma_e | 7.9734117 rho | .47271208 (fraction of variance due to u_i) ------------------------------------------------------------------------------ .
More importantly does this model fit the question I am asking? I read a bunch about the -diff- command but it didn't seem suitable for panel data. Here's to hoping I am on the right track!
Thanks in advance for any guidance!
0 Response to How to interpret results from xtreg difference in differences
Post a Comment