Previous: Why does eiread report

Up: Frequently Asked Questions

Next: How do I make

How do I test whether $B^b-B^w$=0?

You can get simulations of this difference (racially polarized voter turnout in the running example) by doing: aggs=eiread(dbuf,"aggs"); and diff=aggs[.,1]-aggs[.,2];. Then you can use diff as you might normally use simulations of $\beta_i^b$. For example, a point estimate of the aggregate difference is meanc(diff), a standard error is stdc(diff). You can also see the full posterior distribution by plotting the histogram with call hist(diff,7) or a kernel density estimate with call dens(diff).