Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: SPATGSA
From
Maurizio Pisati <[email protected]>
To
[email protected]
Subject
Re: st: SPATGSA
Date
Sun, 31 Oct 2010 07:02:40 +0100
Dear Ivan,
the problem is that your variable of interest (rd_int) contains missing values. Try this:
. use "/Users/Tonzolo/Downloads/moran_working.dta", clear
(Patents granted 1976-2006, incl. ipc and pdpass no)
. drop if rd_int==.
(168 observations deleted)
. su
Variable | Obs Mean Std. Dev. Min Max
-------------+--------------------------------------------------------
state | 343 6 0 6 6
statecounty | 343 6061.659 27.60917 6001 6111
gvkey | 343 10922.32 6591.155 1055 28279
year | 343 1990 0 1990 1990
latitude | 343 35.50675 1.959418 32.90462 39.64835
-------------+--------------------------------------------------------
longitude | 343 -119.8231 2.134776 -122.7686 -116.8036
sic | 343 3501.647 373.6784 2330 3949
rd_int | 343 1.365108 9.319346 0 142.1818
ad_int | 152 .0838932 .5656234 0 6.972574
pat_int | 291 .2141094 1.592589 0 25
. spatwmat , name(wts) xcoord(longitude) ycoord(latitude) band(0, 1000)
The following matrix has been created:
1. Inverse distance weights matrix wts
Dimension: 343x343
Distance band: 0 < d <= 1000
Friction parameter: 1
Minimum distance: 0.0
1st quartile distance: 0.5
Median distance: 3.2
3rd quartile distance: 5.4
Maximum distance: 8.2
Largest minimum distance: 1.53
Smallest maximum distance: 4.26
. spatgsa rd_int, weights(wts) moran
Measures of global spatial autocorrelation
Weights matrix
--------------------------------------------------------------
Name: wts
Type: Distance-based (inverse distance)
Distance band: 0.0 < d <= 1000.0
Row-standardized: No
--------------------------------------------------------------
Moran's I
--------------------------------------------------------------
Variables | I E(I) sd(I) z p-value*
--------------------+-----------------------------------------
rd_int | -0.002 -0.003 0.004 0.168 0.433
--------------------------------------------------------------
*1-tail test
. spatgsa rd_int, weights(wts) geary
Measures of global spatial autocorrelation
Weights matrix
--------------------------------------------------------------
Name: wts
Type: Distance-based (inverse distance)
Distance band: 0.0 < d <= 1000.0
Row-standardized: No
--------------------------------------------------------------
Geary's c
--------------------------------------------------------------
Variables | c E(c) sd(c) z p-value*
--------------------+-----------------------------------------
rd_int | 0.871 1.000 0.268 -0.480 0.315
--------------------------------------------------------------
*1-tail test
Best wishes,
Maurizio
Il giorno 31/ott/2010, alle ore 04.34, Ivan Png ha scritto:
> Greetings!
>
> I'm trying to estimate spatial correlation. I gave the following commands:
>
> use moran_working.dta , clear
> spatwmat , name(wts) xcoord(longitude) ycoord(latitude) band(0, 1000)
> spatgsa rd_int , weights(wts) moran
>
> The result was nothing:
> The following matrix has been created:
>
> 1. Inverse distance weights matrix wts
> Dimension: 511x511
> Distance band: 0 < d <= 1000
> Friction parameter: 1
> Minimum distance: 0.0
> 1st quartile distance: 0.5
> Median distance: 2.8
> 3rd quartile distance: 5.4
> Maximum distance: 10.5
> Largest minimum distance: 2.59
> Smallest maximum distance: 5.79
>
>
>
> . spatgsa rd_int , weights(wts) moran
>
>
> Measures of global spatial autocorrelation
>
>
> Weights matrix
> --------------------------------------------------------------
> Name: wts
> Type: Distance-based (inverse distance)
> Distance band: 0.0 < d <= 1000.0
> Row-standardized: No
> --------------------------------------------------------------
>
> Moran's I
> --------------------------------------------------------------
> Variables | I E(I) sd(I) z p-value*
> --------------------+-----------------------------------------
> rd_int | . -0.002 0.003 . .
> --------------------------------------------------------------
> *1-tail test
>
> ======================
>
>
> I also tried
> spatwmat , name(wts) xcoord(longitude) ycoord(latitude) band(0, 1000)
> spatgsa rd_int , weights(wts) geary
>
> and also got nothing:
>
> The following matrix has been created:
>
> 1. Inverse distance weights matrix wts
> Dimension: 511x511
> Distance band: 0 < d <= 1000
> Friction parameter: 1
> Minimum distance: 0.0
> 1st quartile distance: 0.5
> Median distance: 2.8
> 3rd quartile distance: 5.4
> Maximum distance: 10.5
> Largest minimum distance: 2.59
> Smallest maximum distance: 5.79
>
>
>
> . spatgsa rd_int , weights(wts) geary
>
>
> Measures of global spatial autocorrelation
>
>
> Weights matrix
> --------------------------------------------------------------
> Name: wts
> Type: Distance-based (inverse distance)
> Distance band: 0.0 < d <= 1000.0
> Row-standardized: No
> --------------------------------------------------------------
>
> Geary's c
> --------------------------------------------------------------
> Variables | c E(c) sd(c) z p-value*
> --------------------+-----------------------------------------
> rd_int | . 1.000 0.187 . .
> --------------------------------------------------------------
> *1-tail test
>
> =====================
>
>
> I've had the same problem with hundreds of similar data-sets.
> Grateful for your help. In case it would help in debugging, I've
> uploaded a sample data-set here:
>
> http://www.comp.nus.edu.sg/~ipng/research/moran_working.dta
>
> Many thanks
> Ivan Png
>
>
>
> --
> Skype: ipng00
> T: +65 6516 6807
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/