Why the number of observations is so small?

[Giang Dao]

I used "xtset" to declare my dataset is panel data. The data looks like things below:

code	country	years	var_1	var_2
1	a	2000	...(numeric)	...(numeric)
1	a	2001
2	b	2000
2	b	2001
...	...	...

xtset code year

then I run

xtreg var_1 var_2

and got the number of observations much lower than the one I truly have (number of rows of the table above). Why does STATA drop a lot of observations? Do I have to reshape the table so that the years are unique? If I do this, I lose many observations actually. I just want to see how the regression goes across countries over time, not only for one specific country over time.

Thanks for your help.

[William Lisowski]

This example shows one way of viewing the data for the observations that were dropped. Perhaps that will lead you to some understanding of the problem.

. sysuse auto, clear
(1978 Automobile Data)

. count
  74

. reg price rep78

      Source |       SS           df       MS      Number of obs   =        69
-------------+----------------------------------   F(1, 67)        =      0.00
       Model |  24770.7652         1  24770.7652   Prob > F        =    0.9574
    Residual |   576772188        67  8608540.12   R-squared       =    0.0000
-------------+----------------------------------   Adj R-squared   =   -0.0149
       Total |   576796959        68  8482308.22   Root MSE        =      2934

------------------------------------------------------------------------------
       price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
       rep78 |   19.28012   359.4221     0.05   0.957    -698.1295    736.6897
       _cons |   6080.379    1274.06     4.77   0.000     3537.345    8623.413
------------------------------------------------------------------------------

. list price rep78 if !e(sample)

     +----------------+
     |  price   rep78 |
     |----------------|
  3. |  3,799       . |
  7. |  4,453       . |
 45. |  6,486       . |
 51. |  4,424       . |
 64. | 12,990       . |
     +----------------+

[Phil Bromiley]

To extend William's comment, Stata's regression estimator drops observations that a missing for any of the variables. You might find misstable useful.