Modelling recreational angling demand in Sweden based on region-specific inclusive values

This thesis endeavours to model a trip demand function for recreationalangling in Sweden, including the individual expected per-trip utility of regionalattributes. The analysis is conducted with the use of a Random Utility Model (RUM)for the estimation of ‘site-specific’ utility, and a negative binomial logit model fortrip demand. The site quality variable is stated as expected catch-rates (CR) definedin terms of three different specifications: sample mean CR, individually perfectlyforesighted CR, and an econometrically predicted individual CR. Results indicatethat the econometrically estimated individual catch-rate specification performs wellas an explanatory variable both when modelling discrete site choice and trip demand,while the sample mean and perfectly foresighted CR specifications provide withunintuitive and insignificant parameter values. The inclusive value of the regionchosen by the angler, estimated with the RUM, was subsequently found to be asignificant predicting variable for the number of recreational angling trips conductedby Swedish anglers