Comments on: Coles vs Woolworths – Explaining Sales Differences http://dsanalytics.com/dsblog/coles-vs-woolworths-%e2%80%93-explaining-sales-differences_115 Data Analytics- the art and science of analyzing data Sun, 06 Jul 2008 04:24:05 +0000 http://wordpress.org/?v=2.0.1 by: John Aitchison http://dsanalytics.com/dsblog/coles-vs-woolworths-%e2%80%93-explaining-sales-differences_115#comment-59 Wed, 18 Jul 2007 09:46:52 +0000 http://dsanalytics.com/dsblog/coles-vs-woolworths-%e2%80%93-explaining-sales-differences_115#comment-59 more anecdote http://highriser.blogspot.com/2006/02/coles-v-safeway.html The internet is a great way to develop the concourse, the set of plausible factor, the hypothesis space more anecdote

http://highriser.blogspot.com/2006/02/coles-v-safeway.html

The internet is a great way to develop the concourse, the set of plausible factor, the hypothesis space

]]> by: John Aitchison http://dsanalytics.com/dsblog/coles-vs-woolworths-%e2%80%93-explaining-sales-differences_115#comment-58 Wed, 18 Jul 2007 09:15:27 +0000 http://dsanalytics.com/dsblog/coles-vs-woolworths-%e2%80%93-explaining-sales-differences_115#comment-58 there is some more talk on this issue (mostly anecdotal, addressed to possible plausible explanators rather than to modelling) at http://www.danielbowen.com/2006/03/08/coles-vs-safeway/ and http://diffusionblog.blogspot.com/2006/03/coles-without-myer-still-short-on.html That is of interest, but proposing possible causal factors is a long way from actually getting any quantification, and may even be a retrograde step (the problem then looks very very complex, when it may not be) there is some more talk on this issue (mostly anecdotal, addressed to possible plausible explanators rather than to modelling) at

http://www.danielbowen.com/2006/03/08/coles-vs-safeway/

and

http://diffusionblog.blogspot.com/2006/03/coles-without-myer-still-short-on.html

That is of interest, but proposing possible causal factors is a long way from actually getting any quantification, and may even be a retrograde step (the problem then looks very very complex, when it may not be)

]]>