[R-sig-Geo] Is Fisher z transformation applicable for the z score of Moran's I in function sp.correlogram?

[mozzie]

Dear all,

I'm currious about if Fisher z transformation could be used to determine
that wheter two distinct z score of  Moran's I have statistically
significant difference?

For example, follows are my result :

Spatial correlogram
method: Moran's I
             estimate expectation    variance standard deviate Pr(I) two
sided    
1 (5300)   0.06365873 -0.00018871  0.00017095           4.8833      
1.043e-06 ***
2 (4631)   0.02206724 -0.00021598  0.00022252           1.4938         
0.1352    
3 (3981)  -0.01431569 -0.00025126  0.00027118          -0.8541         
0.3931    
4 (3433)   0.01341846 -0.00029138  0.00033019           0.7545         
0.4506    
5 (2898)   0.02451172 -0.00034518  0.00040564           1.2342         
0.2171    
6 (2393)  -0.00115558 -0.00041806  0.00049807          -0.0330         
0.9736    
7 (1969)   0.01015559 -0.00050813  0.00060344           0.4341         
0.6642    
8 (1630)  -0.00541672 -0.00061387  0.00072303          -0.1786         
0.8582    
9 (1330)   0.01078355 -0.00075245  0.00088182           0.3885         
0.6977    
10 (1094) -0.00027244 -0.00091491  0.00106462           0.0197         
0.9843    
---

Can I use the standard deviate column for Fisher z transformation?



--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/Is-Fisher-z-transformation-applicable-for-the-z-score-of-Moran-s-I-in-function-sp-correlogram-tp7588159.html
Sent from the R-sig-geo mailing list archive at Nabble.com.