我正在寻找您对Stata中var和regress命令之间差异的见解。鉴于相同的变量和相同的滞后数,是什么使这些模型不同(根据其输出的差异来判断)?
var y x1 x2, lags(1/7)
regress L(1/7).y L(1/7).x1 L(1/7).x2
该系列事先被转变为静止。
var y x1 x2, lags(1/7)
Vector autoregression
Sample: 9 - 159 No. of obs = 151
Log likelihood = -2461.622 AIC = 33.47844
FPE = 7.00e+10 HQIC = 34.01421
Det(Sigma_ml) = 2.90e+10 SBIC = 34.79725
Equation Parms RMSE R-sq chi2 P>chi2
---------------------------------------------------------------
y 22 627.086 0.4632 130.3037 0.0000
x1 22 16.4642 0.4150 107.1156 0.0000
x2 22 34.8932 0.3821 93.37647 0.0000
----------------------------------------------------------------
---------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
----------------+----------------------------------------------------------------
y |
y |
L1. | -.8034219 .0870606 -9.23 0.000 -.9740576 -.6327862
L2. | -.829339 .1112633 -7.45 0.000 -1.047411 -.611267
L3. | -.6881525 .1268751 -5.42 0.000 -.9368231 -.4394818
L4. | -.5958702 .1316349 -4.53 0.000 -.8538699 -.3378706
L5. | -.4941909 .1285658 -3.84 0.000 -.7461752 -.2422066
L6. | -.3478784 .1130961 -3.08 0.002 -.5695426 -.1262142
L7. | -.1273106 .0892459 -1.43 0.154 -.3022294 .0476083
|
x1 |
L1. | 2.814694 4.697886 0.60 0.549 -6.392995 12.02238
L2. | 13.40258 5.712821 2.35 0.019 2.205654 24.5995
L3. | 13.41822 6.119334 2.19 0.028 1.424542 25.41189
L4. | 7.634082 6.373183 1.20 0.231 -4.857128 20.12529
L5. | 2.001271 5.898859 0.34 0.734 -9.56028 13.56282
L6. | 3.421364 5.569404 0.61 0.539 -7.494468 14.3372
L7. | 4.068799 4.46953 0.91 0.363 -4.691319 12.82892
|
x2 |
L1. | -.5105249 2.210646 -0.23 0.817 -4.843312 3.822262
L2. | -2.108354 2.495037 -0.85 0.398 -6.998537 2.78183
L3. | -1.442043 2.592775 -0.56 0.578 -6.523789 3.639704
L4. | -.9065004 2.620667 -0.35 0.729 -6.042914 4.229913
L5. | -.0001391 2.53355 -0.00 1.000 -4.965806 4.965528
L6. | 2.146481 2.427015 0.88 0.376 -2.610381 6.903343
L7. | -1.118613 2.118762 -0.53 0.598 -5.271309 3.034084
|
_cons | 22.43668 48.04635 0.47 0.641 -71.73243 116.6058
----------------+----------------------------------------------------------------
x1 |
y |
L1. | .0036968 .0022858 1.62 0.106 -.0007833 .0081768
L2. | .0012158 .0029212 0.42 0.677 -.0045097 .0069413
L3. | .0035081 .0033311 1.05 0.292 -.0030208 .010037
L4. | .0032596 .0034561 0.94 0.346 -.0035142 .0100334
L5. | .0005852 .0033755 0.17 0.862 -.0060307 .007201
L6. | -.0018743 .0029693 -0.63 0.528 -.0076941 .0039455
L7. | -.0040389 .0023432 -1.72 0.085 -.0086314 .0005537
|
x1 |
L1. | -.5753736 .1233434 -4.66 0.000 -.8171223 -.3336249
L2. | -.3020477 .1499906 -2.01 0.044 -.5960239 -.0080714
L3. | -.3313213 .1606637 -2.06 0.039 -.6462164 -.0164263
L4. | -.1718872 .1673285 -1.03 0.304 -.4998451 .1560707
L5. | -.1834757 .1548751 -1.18 0.236 -.4870253 .1200739
L6. | .0489376 .1462252 0.33 0.738 -.2376586 .3355337
L7. | .1766427 .1173479 1.51 0.132 -.053355 .4066404
|
x2 |
L1. | -.1051509 .0580407 -1.81 0.070 -.2189086 .0086069
L2. | -.1006968 .0655074 -1.54 0.124 -.229089 .0276954
L3. | -.0906552 .0680736 -1.33 0.183 -.2240769 .0427665
L4. | -.1436015 .0688059 -2.09 0.037 -.2784585 -.0087445
L5. | -.0930764 .0665186 -1.40 0.162 -.2234505 .0372976
L6. | -.1018913 .0637215 -1.60 0.110 -.2267832 .0230006
L7. | -.1194924 .0556283 -2.15 0.032 -.2285218 -.0104629
|
_cons | 1.918878 1.261461 1.52 0.128 -.553541 4.391296
----------------+----------------------------------------------------------------
x2 |
y |
L1. | .0010281 .0048444 0.21 0.832 -.0084667 .0105228
L2. | -.0038838 .0061911 -0.63 0.530 -.0160181 .0082505
L3. | .0035605 .0070598 0.50 0.614 -.0102764 .0173974
L4. | .0041767 .0073246 0.57 0.569 -.0101793 .0185327
L5. | .0007593 .0071538 0.11 0.915 -.013262 .0147806
L6. | -.0027897 .0062931 -0.44 0.658 -.0151239 .0095445
L7. | .0018272 .004966 0.37 0.713 -.0079059 .0115603
|
x1 |
L1. | .3332696 .2614066 1.27 0.202 -.179078 .8456172
L2. | .6160613 .3178811 1.94 0.053 -.0069742 1.239097
L3. | .4139762 .3405009 1.22 0.224 -.2533934 1.081346
L4. | .2837896 .3546259 0.80 0.424 -.4112645 .9788436
L5. | .4448436 .3282329 1.36 0.175 -.1984811 1.088168
L6. | .6417029 .3099009 2.07 0.038 .0343084 1.249098
L7. | .4719593 .2487001 1.90 0.058 -.0154839 .9594025
|
x2 |
L1. | -.7465681 .123008 -6.07 0.000 -.9876594 -.5054769
L2. | -.6760273 .1388325 -4.87 0.000 -.948134 -.4039206
L3. | -.4367948 .144271 -3.03 0.002 -.7195607 -.1540289
L4. | -.4889316 .145823 -3.35 0.001 -.7747393 -.2031238
L5. | -.5310379 .1409755 -3.77 0.000 -.8073447 -.254731
L6. | -.4416263 .1350475 -3.27 0.001 -.7063146 -.1769381
L7. | -.3265204 .1178952 -2.77 0.006 -.5575907 -.09545
|
_cons | 3.568261 2.673465 1.33 0.182 -1.671634 8.808155
---------------------------------------------------------------------------------
regress L(1/7).y L(1/7).x1 L(1/7).x2
Source | SS df MS Number of obs = 151
-------------+------------------------------ F( 20, 130) = 7.23
Model | 49291082.3 20 2464554.11 Prob > F = 0.0000
Residual | 44322342.8 130 340941.099 R-squared = 0.5265
-------------+------------------------------ Adj R-squared = 0.4537
Total | 93613425.1 150 624089.501 Root MSE = 583.9
---------------------------------------------------------------------------------
L.y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
----------------+----------------------------------------------------------------
y |
L2. | -.8074369 .0868829 -9.29 0.000 -.9793244 -.6355494
L3. | -.7857941 .1076428 -7.30 0.000 -.9987525 -.5728357
L4. | -.6747462 .1186733 -5.69 0.000 -.9095271 -.4399654
L5. | -.5758927 .1192639 -4.83 0.000 -.811842 -.3399433
L6. | -.4199846 .1078154 -3.90 0.000 -.6332845 -.2066846
L7. | -.2444889 .0873128 -2.80 0.006 -.4172267 -.071751
|
x1 |
L1. | 9.174249 4.663798 1.97 0.051 -.0525176 18.40102
L2. | 6.026435 5.730833 1.05 0.295 -5.311334 17.3642
L3. | 13.03098 6.057813 2.15 0.033 1.046324 25.01564
L4. | 13.01178 6.318175 2.06 0.041 .5120225 25.51153
L5. | 6.146548 5.91807 1.04 0.301 -5.561646 17.85474
L6. | .8687361 5.610159 0.15 0.877 -10.23029 11.96776
L7. | -.6015264 4.502342 -0.13 0.894 -9.508873 8.30582
|
x2 |
L1. | 2.709283 2.214315 1.22 0.223 -1.671474 7.090041
L2. | 2.947753 2.500195 1.18 0.241 -1.998585 7.89409
L3. | .7449778 2.611172 0.29 0.776 -4.420914 5.910869
L4. | .8159876 2.639117 0.31 0.758 -4.405191 6.037166
L5. | 1.839693 2.54722 0.72 0.471 -3.199677 6.879062
L6. | 2.267241 2.436901 0.93 0.354 -2.553876 7.088358
L7. | 4.198018 2.102467 2.00 0.048 .0385389 8.357497
|
_cons | -3.078699 48.40164 -0.06 0.949 -98.83556 92.67816
---------------------------------------------------------------------------------
答案 0 :(得分:0)
对我来说,他们有两种不同的规格。
第一个(VAR)正在估计三个自变量滞后对因变量(y,x1,x2)的影响。第二个是估计Lag从2:7的y + Lags对x1 + Lags(1:7)的x2 + Lags对因变量L(y)的影响。所以他们在y方面有两个不同的因变量和自变量。请参阅下面的等式(前三个用于var
代码,最后一个用于regress
代码):
OLS规范未考虑模型中变量之间存在的反馈效果。虽然你可能对X1对y的影响感兴趣,但X1也会受到y及其滞后值 - 反馈效应的影响。因此,使用OLS将导致虚假回归。