We know "reg i.id i.year,vce(cl id)" is the same as "xtreg i.year,fe r",but why are the results of the two codes are different in standard error and t statistics.Here is the example.
Code:
. reg invest mvalue kstock i.company i.year,vce(cl company)
Linear regression Number of obs = 200
F(8, 9) = .
Prob > F = .
R-squared = 0.9517
Root MSE = 51.725
(Std. Err. adjusted for 10 clusters in company)
------------------------------------------------------------------------------
| Robust
invest | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mvalue | 0.118 0.011 10.60 0.000 0.093 0.143
kstock | 0.358 0.049 7.29 0.000 0.247 0.469
|
company |
2 | 207.054 39.027 5.31 0.000 118.769 295.339
3 | -135.231 35.132 -3.85 0.004 -214.704 -55.758
4 | 95.354 59.423 1.60 0.143 -39.070 229.778
5 | -5.439 50.629 -0.11 0.917 -119.969 109.092
6 | 102.889 62.961 1.63 0.137 -39.539 245.316
7 | 51.467 57.614 0.89 0.395 -78.864 181.798
8 | 67.490 61.119 1.10 0.298 -70.770 205.751
9 | 30.218 56.268 0.54 0.604 -97.070 157.505
10 | 126.837 70.615 1.80 0.106 -32.905 286.579
|
year |
1936 | -19.197 21.243 -0.90 0.390 -67.251 28.857
1937 | -40.690 34.158 -1.19 0.264 -117.961 36.581
1938 | -39.226 16.150 -2.43 0.038 -75.760 -2.692
1939 | -69.470 27.708 -2.51 0.033 -132.151 -6.790
1940 | -44.235 17.829 -2.48 0.035 -84.567 -3.903
1941 | -18.804 18.317 -1.03 0.331 -60.239 22.630
1942 | -21.140 14.537 -1.45 0.180 -54.025 11.745
1943 | -42.978 12.874 -3.34 0.009 -72.100 -13.855
1944 | -43.099 11.285 -3.82 0.004 -68.627 -17.571
1945 | -55.683 15.601 -3.57 0.006 -90.976 -20.390
1946 | -31.169 21.467 -1.45 0.180 -79.730 17.392
1947 | -39.392 27.132 -1.45 0.180 -100.769 21.984
1948 | -43.717 39.900 -1.10 0.302 -133.978 46.544
1949 | -73.495 39.260 -1.87 0.094 -162.307 15.317
1950 | -75.896 37.766 -2.01 0.075 -161.328 9.536
1951 | -62.481 50.717 -1.23 0.249 -177.211 52.249
1952 | -64.632 52.917 -1.22 0.253 -184.339 55.075
1953 | -67.718 44.894 -1.51 0.166 -169.276 33.840
1954 | -93.526 32.560 -2.87 0.018 -167.182 -19.870
|
_cons | -86.900 62.010 -1.40 0.195 -227.178 53.377
------------------------------------------------------------------------------
. xtreg invest mvalue kstock i.year,fe r
Fixed-effects (within) regression Number of obs = 200
Group variable: company Number of groups = 10
R-sq: Obs per group:
within = 0.7985 min = 20
between = 0.8143 avg = 20.0
overall = 0.8068 max = 20
F(9,9) = .
corr(u_i, Xb) = -0.3250 Prob > F = .
(Std. Err. adjusted for 10 clusters in company)
------------------------------------------------------------------------------
| Robust
invest | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mvalue | 0.118 0.011 10.88 0.000 0.093 0.142
kstock | 0.358 0.048 7.48 0.000 0.250 0.466
|
year |
1936 | -19.197 20.699 -0.93 0.378 -66.021 27.626
1937 | -40.690 33.283 -1.22 0.253 -115.982 34.602
1938 | -39.226 15.736 -2.49 0.034 -74.825 -3.628
1939 | -69.470 26.999 -2.57 0.030 -130.546 -8.395
1940 | -44.235 17.372 -2.55 0.031 -83.534 -4.936
1941 | -18.804 17.847 -1.05 0.320 -59.178 21.569
1942 | -21.140 14.165 -1.49 0.170 -53.183 10.903
1943 | -42.978 12.544 -3.43 0.008 -71.354 -14.601
1944 | -43.099 10.996 -3.92 0.004 -67.973 -18.224
1945 | -55.683 15.202 -3.66 0.005 -90.072 -21.294
1946 | -31.169 20.917 -1.49 0.170 -78.487 16.148
1947 | -39.392 26.437 -1.49 0.170 -99.197 20.413
1948 | -43.717 38.879 -1.12 0.290 -131.666 44.233
1949 | -73.495 38.254 -1.92 0.087 -160.033 13.043
1950 | -75.896 36.798 -2.06 0.069 -159.140 7.348
1951 | -62.481 49.418 -1.26 0.238 -174.272 49.311
1952 | -64.632 51.562 -1.25 0.242 -181.274 52.009
1953 | -67.718 43.745 -1.55 0.156 -166.675 31.239
1954 | -93.526 31.726 -2.95 0.016 -165.296 -21.756
|
_cons | -32.836 19.783 -1.66 0.131 -77.588 11.915
-------------+----------------------------------------------------------------
sigma_u | 91.798268
sigma_e | 51.724523
rho | .75902159 (fraction of variance due to u_i)
------------------------------------------------------------------------------Many thanks in advance.
Raymond
0 Response to reg i.id i.year,vce(cl id) vs. xtreg i.year,fe r
Post a Comment