> To: Subject: Re: st: RE: RE: RE: AW: Creating index relative to other observations If I just stack observations type i on top of observations type j, geodist doesn't like the missing values (observations type i have missing values for type j, and vice versa). For each location of type i, I need to compute the distances to every location of type j.
#GEODIST TWO DATASETS CODE#
> Robert Picard sent the code below, which works as advertised - many thanks, Robert! Now I have a slightly different problem: I have two kinds of locations in the data, i and j. Suppose you want the weighted sum of inverseĭistances (assuming none are zero) then you could just: Type j, you should not create that many new variables at once, toĬonserve memory. Store distances if you only want summary stats across locations of Of type j stored on the type i obs, you will need Nj new variables to Suppose your location variables are xi,yi,xj,yj and you have Ni obs of Variable names must be distinct across the two datasets. Strategy (merging on _n rather than any variables) is that all the The crucial detail to remember with an unmatched merge Single observation from the first, but this is understandably lessĮfficient. Published 2009 in the JHE-see also Appendix A ofĭistance weights) as a way to have two datasets in memory at once Īnother way is to repeatedly merge or append a second dataset onto a Name came later the approach was developed in 2003 for a paper Package for any of those spherical approximate computations seems likeĪs far as I know, the unmatched merge approach was first promulgated Plus, the formula in that message is the weakest of many alternatives Lat/lon occupies a large fraction of msg00473's code, but need not Īll the calculations could be telescoped into one line (it's justĮasier to break it up), and the local macros are mostly unnecessary. Note also the calculation of distance (using an approximation thatĪssumes the Earth is a sphere see -vincenty- on SSC for anĪlternative) between two points on Earth measured in decimal degrees There are numerous examples addressing your need in the Archives, e.g.: In our case we don't have a prior measure of anything comparable to the market share used in your paper, so we can't use the nice trick you used there. The discussion of inverse distance indices in Horwitz & Nichols is very useful as well. Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at.
0 kommentar(er)
