increasing obs per group in data panel regression

[Joe Mckennedy]

Hello I was wondering if the number of obs is 324 and number of groups is 325. Why do I get only 8 or 9 obs per group in my panel data regression?

. xtreg log_price unemployment mortgage_rate log_income ,fe cluster(ID_code)

Fixed-effects (within) regression Number of obs = 2594
Group variable: ID_code Number of groups = 325

R-sq: Obs per group:
within = 0.7181 min = 7
between = 0.6936 avg = 8.0
overall = 0.2121 max = 8

F(4,324) = 2066.76
corr(u_i, Xb) = -0.6932 Prob > F = 0.0000

(Std. Err. adjusted for 325 clusters in GM_code)
-----------------------------------------------------------------------------------
| Robust
log_price | Coef. Std. Err. t P>|t| [95% Conf. Interval]
------------------+----------------------------------------------------------------
unemployment | .0962734 .0061217 15.73 0.000 .0842301 .1083167
mortgage_rate | .0049352 .0011974 4.12 0.000 .0025796 .0072908
log_income | -.3595322 .0633615 -5.67 0.000 -.4841841 -.2348802
_cons | 1.266938 .360273 3.52 0.000 .558168 1.975707
------------------+----------------------------------------------------------------
sigma_u | .28079617
sigma_e | .03753937
rho | .98244106 (fraction of variance due to u_i)
-----------------------------------------------------------------------------------

[Jared Greathouse]

Because 2594/8=324.

How many observations per group did you think you'd have?

[Joe Mckennedy]

Because 2594/8=324.

How many observations per group did you think you'd have?

I mean having 8 obs per group for example it doesnt make it biased?
Also why is unemployment and mortgage rate positive, if I run normal regression they are both negative?

[Jared Greathouse]

Why would it make it biased?

And I don't know, different estimators will yield different results, especially when you used fixed effects estimators and additional covariates