Mikko M�kinen wrote:
> I'm also a new user of Stata :-) As Justin wrote, you can calculate the
> D-W statistic by using dummies and OLS. So, how does the -dwstat2- command
> differ from the dummies and OLS approach?
Apparently, not much at all. Using wagepan.dta from Jeffrey M.
Wooldridge's suite of datasets
(http://wooldridge-datasets.swlearning.com/):
. gen id = _n
. tsset id
time variable: id, 1 to 4360
. reg lwage year agric- union expersq
Source | SS df MS Number of obs =
4360
-------------+------------------------------ F( 33, 4326) =
52.71
Model | 354.627996 33 10.7463029 Prob > F =
0.0000
Residual | 881.901646 4326 .20386076 R-squared =
0.2868
-------------+------------------------------ Adj R-squared =
0.2814
Total | 1236.52964 4359 .283672779 Root MSE =
.45151
------------------------------------------------------------------------------
lwage | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
-------------+----------------------------------------------------------------
year | .0175141 .0062004 2.82 0.005 .0053582
.02967
agric | .2171099 .0760737 2.85 0.004 .0679665
.3662534
black | -.1226481 .0229259 -5.35 0.000 -.1675947
-.0777015
bus | .3931705 .0620011 6.34 0.000 .2716166
.5147245
construc | .4576362 .0628484 7.28 0.000 .3344212
.5808513
ent | (dropped)
exper | .0603432 .0118913 5.07 0.000 .0370301
.0836564
fin | .5605775 .0676363 8.29 0.000 .4279758
.6931792
hisp | -.0413889 .0217987 -1.90 0.058 -.0841256
.0013478
poorhlth | -.0304731 .0532302 -0.57 0.567 -.1348315
.0738854
hours | -.0000874 .0000131 -6.70 0.000 -.000113
-.0000618
manuf | .5456511 .0588261 9.28 0.000 .4303219
.6609804
married | .1130483 .0152518 7.41 0.000 .083147
.1429495
min | .7033441 .0803993 8.75 0.000 .5457203
.860968
nrthcen | -.095112 .0223457 -4.26 0.000 -.138921
-.051303
nrtheast | .0214095 .0230724 0.93 0.353 -.0238242
.0666431
occ1 | .0854797 .0387491 2.21 0.027 .0095117
.1614477
occ2 | .0689334 .03759 1.83 0.067 -.0047622
.142629
occ3 | (dropped)
occ4 | -.0860991 .036848 -2.34 0.020 -.15834
-.0138582
occ5 | -.0264525 .0354058 -0.75 0.455 -.0958659
.0429609
occ6 | -.1017852 .0361376 -2.82 0.005 -.1726333
-.0309371
occ7 | -.1267472 .0394001 -3.22 0.001 -.2039916
-.0495029
occ8 | -.0688939 .0801958 -0.86 0.390 -.2261188
.0883311
occ9 | -.1451409 .0378897 -3.83 0.000 -.2194242
-.0708577
per | .388633 .077212 5.03 0.000 .2372579
.5400081
pro | .2148503 .0615476 3.49 0.000 .0941854
.3355151
pub | .4359741 .0661164 6.59 0.000 .3063521
.5655961
rur | -.110404 .0184424 -5.99 0.000 -.1465605
-.0742474
south | -.0791108 .0207352 -3.82 0.000 -.1197624
-.0384593
educ | .0742839 .0052508 14.15 0.000 .0639897
.0845781
tra | .5766668 .0631995 9.12 0.000 .4527635
.7005701
trad | .3201914 .058149 5.51 0.000 .2061896
.4341932
union | .173667 .0169976 10.22 0.000 .1403431
.206991
expersq | -.0019649 .0006943 -2.83 0.005 -.003326
-.0006038
_cons | -34.43747 12.23924 -2.81 0.005 -58.43265
-10.44229
------------------------------------------------------------------------------
. dwstat
Durbin-Watson d-statistic( 34, 4360) = 2.096452
. dwstat2
Durbin-Watson d-statistic( 34, 4360) = 2.096452
With apologies for the messy output. Kit Baum may be able to shed more
light on what the "minor modifications" are for -dwstat2- when he has time
to reply.
CLIVE NICHOLAS |t: 0(044)191 222 5969
Politics |e: [email protected]
Newcastle University |http://www.ncl.ac.uk/geps
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
