Estimates fixed effects models:
y_{it} = a + b x_{it} + c_{i}. It cannot be abbreviated.

**Typical Use:**

`areg depvar indvar1 indvar2, absorb(groupvar)`

. areg hhsize age, a(ethnicity) Number of obs = 1081 F( 1, 1067) = 16.35 Prob > F = 0.0001 R-squared = 0.0510 Adj R-squared = 0.0395 Root MSE = 4.5769 ------------------------------------------------------------------------------ hhsize | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | .0367759 .0090946 4.04 0.000 .0189305 .0546213 _cons | 5.974907 .4004147 14.92 0.000 5.189218 6.760597 -------------+---------------------------------------------------------------- ethnicity | F(12, 1067) = 3.458 0.000 (13 categories)Here, we estimate the effect of age on household size, controlling for ethnicity. We see that, even controlling for ethnicity, older respondents have larger households. Overall, household sizes are significantly different for different ethnicities (the F-test in the last line tells us this).

- The option
`, r`makes Stata calculate heteroskedastic robust standard errors.

- There are many ways to implement a fixed effects model in Stata, see
notes on fixed effects models
for details.
`areg`is almost always the easiest, and supresses the values of the coefficients for each dummy variable while automatically calculating and displaying the test of their joint significance. `areg`is nice, but aareg is nicer -- it allows you to absorb multiple group variables.- See the official Stata help for
`areg`.

contact: djiboliz@gmail.com

last modified: 26 April 2007