How to run LM- tes to check of Spatial Auto-correlation in panel data?

[Abdo Ali]

Hello Dears

I had run LM- tes to check of Spatial Auto-correlation in panel data but i got an error massage

I followed this steeps :

. use "G:\major\PhD\DATA\DATAfinal\datapaper2\data 4-12-2019.dta", clear

. xtset id year
panel variable: id (strongly balanced)
time variable: year, 1992 to 2018
delta: 1 unit

. spatwmat using G:\major\PhD\DATA\DATAfinal\datapaper2\w1.dta, name(W)


The following matrix has been created:

1. Imported non-binary weights matrix W
Dimension: 57x57



. regress lpm10 lfdipc lgdppc lencpc lecbpc lecffpc,

Source | SS df MS Number of obs = 1,539
-------------+---------------------------------- F(5, 1533) = 42.98
Model | 151.987114 5 30.3974228 Prob > F = 0.0000
Residual | 1084.13059 1,533 .707195428 R-squared = 0.1230
-------------+---------------------------------- Adj R-squared = 0.1201
Total | 1236.11771 1,538 .803717624 Root MSE = .84095

------------------------------------------------------------------------------
lpm10 | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lfdipc | -.0220951 .0248719 -0.89 0.374 -.0708817 .0266914
lgdppc | .0550906 .0861201 0.64 0.522 -.113835 .2240163
lencpc | -.6031066 .1194508 -5.05 0.000 -.8374108 -.3688023
lecbpc | .8812698 .0875119 10.07 0.000 .709614 1.052926
lecffpc | .0367151 .0726866 0.51 0.614 -.1058607 .1792908
_cons | 3.087118 .3380565 9.13 0.000 2.424016 3.75022
------------------------------------------------------------------------------

when I ren this command i got this massage:

. spatdiag, weights(W)
Matrix W is 57x57, regression has been carried out on 1539 obs.
To run -spatdiag- weights matrix dimension must equal N. of obs

pleas could u help me to solve this problem?
Thanks in advance

[Arogundade Sodiq]

Hello Abdo Ali, I want to ask if you have solved this problem. I am also having a similar issue.

Please how did you solve it.

I look forward to hearing from you

[jamal mamkhezri]

That occurs because your weight matrix has different dimensions than your panel data. To solve this, you should restrict your spatdiag command by year. You could try what follows.

forvalues i=1992/2018 {
preserve
keep if year==`i'
regress lpm10 lfdipc lgdppc lencpc lecbpc lecffpc
spatdiag,weights(W)
restore
}

You may want to read Stata's spmatrix help document, pages 108 and on: Here