[Openspace] two sided significance test
anselin at uiuc.edu
Tue Dec 2 14:42:13 CST 2003
The permutation tests are one-sided. The pseudo significance
is (M + 1) / (R + 1) where R is the number of replications and
M is the number of instances where a statistic computed from the
permutations is equal to or greater than the observed (for positive
Moran) or less or equal to the observed (for negative local Moran).
In other words, equal to or more "extreme", which is one-sided.
On Tuesday, December 2, 2003, at 02:21 PM, ccm01 at health.state.ny.us
> This is a note asking for clarification/confirmation on the methods
> used in
> GeoDa. Is the significance filter in GeoDa based on a two-sided test?
> is, in order to get an alpha of 0.05, does GeoDa use the top 2.5% and
> bottom 2.5% of permutation results to determine statistical
> Colleen McLaughlin, MPH, CTR
> Research Scientist
> New York State Cancer Registry, NYSDOH
> voice (518) 474-2255
> fax (518) 473-6789
> ccm01 at health.state.ny.us
> Openspace mailing list
> Openspace at sal.agecon.uiuc.edu
More information about the Openspace