help spmat summarize-------------------------------------------------------------------------------

Title

spmat summarize-- Summarize the spatial-weighting matrixWcontained in anspmatobject

Syntax

spmatsummarizeobjname[,options]

optionsDescription -------------------------------------------------------------------------linksreport linksdetaildisplay detailed summary of linksbandeddisplay banded informationbtruncate(b B)bin truncationdtruncate(l u)diagonal truncationvtruncate(#)value truncation ------------------------------------------------------------------------- Only one ofbanded,btruncate(),dtruncate(), orvtruncate()may be specified.

Description

spmat summarizesummarizes the spatial-weighting matrixWcontained in thespmatobjectobjname. If the object contains annxnmatrix, the user can optionally request summary statistics that show whether the matrix can be stored in a banded form.

Options

linksrequests thatspmat summarizereports links rather than the default values. This option is useful for summarizing contiguity matrices.

detailrequests a detailed tabulation of links.detailwill also display IDs of observations with the minimum and maximum number of links.

bandedreports the bands for the matrix that has already a (possibly) banded structure but is stored in annxnform.

truncate()options report the bands for the matrix after its values have been changed to zero according to one of the truncation criterion.

btruncate(b B)partitions the values ofWintoBbins and truncates to zero those entries that fall into binbor below.

dtruncate(l u)truncates to zero the values ofWthat fallldiagonals below andudiagonals above the main diagonal. Neither value can be greater thanfloor((cols(W)-1)/4).

vtruncate(#)truncates to zero the values ofWthat are less than or equal to #.

ExampleSetup

. spmat use cobj using pollute.spmatSummarize the contiguity matrix contained in the spmat object

cobj. spmat summarize cobj, links

spmat summarizesaves the following inr():Scalars

r(b)number of rows inWr(n)number of columns inWr(lband)lower band ofWr(uband)upper band ofWr(min)minimum ofWr(min0)minimum element > 0 inWr(mean)mean ofWr(max)maximum ofWr(lmin)minimum number of neighbors inWr(lmean)mean number of neighbors inWr(lmax)maximum number of neighbors inWr(ltotal)total number of neighbors inWr(eig)1if object contains eigenvalues,0otherwiser(canband)1ifWcan be banded based onr(lband)andr(uband),0otherwise

Also seeOnline:

spmat,spreg,spivreg,spmap,shp2dta,mif2dta(if installed)