Dears
I try to calibrate a model, model required to estimate potential stores
turnover. Here, I need to calibrate the parameter K, in order to define a
probability of visit from different geographic zones to a single store.
This coeff. K is used in the following formula:
P = A / D Exp (K)
where
P = probability of visit from a single geographic zone to a single store
A = attractivity index of a store
D = distance in minue betweem a single geographic zone and a single store.
K = coefficient of decay related to the distance (always superior to zero)
Data are issued from a survey organized in a store, during a full week. 915
respondants, representative of average store consumer. Error range: 3,2%
Geographic origin of consumers have been geocoded (field zone_house) giving us
a time distance in minute between consumers and store selected.
I would require, if you find you could do this work, to write a methodology to
process this issue, so I could make some more works of this kind in the near
future.
Please let me know whether you'd be interested to process this work, your
charges for that and the time frame for its realisation. I will send you the
set of data then to perform the work.
Best regards
Arnaud
arnaudtrous...@hotmail.com
[[alternative HTML version deleted]]
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
