[求助]STATA中如何检验面板数据的异方差、和自相关性？

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

<P align=center>我在做面板数据模型时，模型的R2值很高，但是各个变量系数的P值也很大，请问这是什么原因？是不是存在异方差或者自相关性引起的？该如何检验？</P>

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏7 回帖

关键词：Stata 自相关性 面板数据 tata 异方差 相关性 如何 模型

用xtgls估计，加上panels(he)和corr(psar1)即可既控制异方差又可消除序列相关的影响。

xtgls c d e f g h i j k l m n o p q r s t u v w x y z aa ab ac

Cross-sectional time-series FGLS regression

Coefficients: generalized least squares

Panels: homoskedastic

Correlation: no autocorrelation

Estimated covariances = 1 Number of obs = 36

Estimated autocorrelations = 0 Number of groups = 4

Estimated coefficients = 27 Time periods = 9

Wald chi2(26) = 5773.39

Log likelihood = 81.31917 Prob > chi2 = 0.0000

------------------------------------------------------------------------------

c | Coef. Std. Err. z P>|z| [95% Conf. Interval]

-------------+----------------------------------------------------------------

d | -7.095139 2.749709 -2.58 0.010 -12.48447 -1.705808

e | .4718551 .5244086 0.90 0.368 -.5559668 1.499677

f | -3.962355 1.562304 -2.54 0.011 -7.024415 -.900295

g | -.7499423 1.49499 -0.50 0.616 -3.680069 2.180184

h | -.0721934 .0268321 -2.69 0.007 -.1247834 -.0196034

i | 1.093924 .3015136 3.63 0.000 .5029678 1.684879

j | -.3219019 .0993785 -3.24 0.001 -.5166802 -.1271236

k | .2802148 .0901929 3.11 0.002 .10344 .4569897

l | .8650244 .1634246 5.29 0.000 .544718 1.185331

m | -.3017405 .1538762 -1.96 0.050 -.6033322 -.0001488

n | .0097719 .0038048 2.57 0.010 .0023146 .0172293

o | .0321919 .1551508 0.21 0.836 -.2718981 .3362818

p | .3218549 .088089 3.65 0.000 .1492036 .4945063

q | -.3220883 .1487057 -2.17 0.030 -.613546 -.0306305

r | .0958412 .1439849 0.67 0.506 -.186364 .3780464

s | -.1332638 .0715016 -1.86 0.062 -.2734044 .0068768

t | .0011726 .0109285 0.11 0.915 -.0202469 .022592

u | -.0044593 .0100168 -0.45 0.656 -.0240917 .0151732

v | -.0004442 .0142002 -0.03 0.975 -.0282762 .0273877

w | .007534 .0150099 0.50 0.616 -.0218849 .036953

x | -.0021665 .010227 -0.21 0.832 -.022211 .017878

y | -.0087039 .007757 -1.12 0.262 -.0239073 .0064994

z | .0302018 .0122591 2.46 0.014 .0061744 .0542291

aa | .0269482 .008834 3.05 0.002 .0096339 .0442625

ab | -.0676077 .0104505 -6.47 0.000 -.0880903 -.0471252

ac | -.0066055 .0130977 -0.50 0.614 -.0322766 .0190656

_cons | 39.36531 13.53426 2.91 0.004 12.83865 65.89196

xtgls c d e f g h i j k l m n o p q r s t u v w x y z aa ab ac,panels(he) corr(psar1)

Cross-sectional time-series FGLS regression

Coefficients: generalized least squares

Panels: heteroskedastic

Correlation: panel-specific AR(1)

Estimated covariances = 4 Number of obs = 36

Estimated autocorrelations = 4 Number of groups = 4

Estimated coefficients = 27 Time periods = 9

Wald chi2(26) = 8734.12

Log likelihood = 84.04154 Prob > chi2 = 0.0000

------------------------------------------------------------------------------

c | Coef. Std. Err. z P>|z| [95% Conf. Interval]

-------------+----------------------------------------------------------------

d | -6.651209 2.599695 -2.56 0.011 -11.74652 -1.555901

e | .6910577 .493424 1.40 0.161 -.2760357 1.658151

f | -2.885732 1.47413 -1.96 0.050 -5.774974 .0035101

g | -1.551059 1.475064 -1.05 0.293 -4.442132 1.340013

h | -.0675715 .0270832 -2.49 0.013 -.1206536 -.0144895

i | 1.049269 .2827947 3.71 0.000 .4950017 1.603536

j | -.3013187 .0886761 -3.40 0.001 -.4751206 -.1275168

k | .2409156 .0826609 2.91 0.004 .0789032 .402928

l | .762575 .1619813 4.71 0.000 .4450975 1.080053

m | -.3457759 .1479114 -2.34 0.019 -.6356768 -.0558749

n | .0096737 .0036398 2.66 0.008 .0025398 .0168076

o | -.0135502 .1435777 -0.09 0.925 -.2949573 .2678569

p | .2825038 .0836345 3.38 0.001 .1185831 .4464245

q | -.2537533 .1452435 -1.75 0.081 -.5384254 .0309188

r | .1402158 .1403893 1.00 0.318 -.1349421 .4153737

s | -.1098276 .0672562 -1.63 0.102 -.2416473 .0219921

t | .0030314 .0104625 0.29 0.772 -.0174747 .0235374

u | -.005772 .0103738 -0.56 0.578 -.0261042 .0145603

v | -.00242 .0135296 -0.18 0.858 -.0289376 .0240975

w | .0022872 .0139579 0.16 0.870 -.0250698 .0296442

x | -.0015213 .0092208 -0.16 0.869 -.0195938 .0165512

y | -.0033703 .0077328 -0.44 0.663 -.0185264 .0117858

z | .0323861 .0121387 2.67 0.008 .0085947 .0561774

aa | .0291652 .0081208 3.59 0.000 .0132488 .0450816

ab | -.0655527 .0099574 -6.58 0.000 -.0850688 -.0460366

ac | -.0031798 .013152 -0.24 0.809 -.0289571 .0225976

_cons | 34.65713 12.76573 2.71 0.007 9.636746 59.67751

谁能告诉我这两个之间的区别？哪个更好？

在论坛搜索我的文章,我也贴了几个相关程序

楼上的我没有足够的MONEY，可不可以先给我发一份？我有些计量经济学等方面的资料，可以交换不？邮箱：pengminling@yahoo.com.cn

maybe

. xttest1, unadjusted

last estimates not found
r(301);这是什么问题啊？就帮助

zgf0910 发表于 2007-6-8 16:47
用xtgls估计，加上panels(he)和corr(psar1)即可既控制异方差又可消除序列相关的影响。

谢谢分享

zgf0910 发表于 2007-6-8 16:47
用xtgls估计，加上panels(he)和corr(psar1)即可既控制异方差又可消除序列相关的影响。

谢谢，正在被这方面的问题困扰