AGNELLO: ECONOMIC VALUE OF FISHING SUCCESS 



previously. The variables in the Z vector are defined 

 in Tables 2 and 3. These variables reflect the addi- 

 tive and interactive (multiplicative) dummy variables 

 which allow us to test for parameter differences 

 across target species. Since the control group in all 

 regressions is bluefish (i.e., anglers indicating blue- 

 fish as the species preference), qualitative (0,1) vari- 

 ables for flounder (F) and weakfish (W), along with 

 their interactions with other exogenous variables are 

 included in each regression. F tests (noted as F 

 (species) in Table 2 and 3) were performed on the 

 interaction and additive dummy variable terms. For 

 the demand frequency regressions (Table 2), since 

 the F (species) statistics for both the additive and 

 multiplicative terms are insignificant, the data can 

 be combined across target species. Thus, model (1) 

 for both OLS and WLS are most appropriate when 

 using Table 2). In the demand price regressions 

 (Table 3), the species terms have significant F- 



statistics (to at least the 0.05 level) indicating that 

 intercept and slope coefficients are different across 

 species. Thus, models OLS (3) and WLS (3) are most 

 appropriate from Table 3. 



The empirical findings for the demand price model 

 (Table 3) are stronger than for the demand frequen- 

 cy model (Table 2) although both have significant 

 equation F-statistics (probability values < 0.05). 

 WLS increases the significance of the results in 

 Table 3 but lowers significance levels in Table 2. The 

 parameter estimates for the travel cost and frequen- 

 cy coefficients (bj and aj < 0) as well as the success 

 coefficients (bg and a, > 0) generally confirm theo- 

 retical expectations. Travel cost and frequency are 

 significantly inversely related, and fishing success 

 as measured by the number of fish kept is general- 

 ly a significant determinant of both fishing fre- 

 quency and travel distance. Various measures and 

 combinations of fishing success were investigated. 



Table 3.— Log-linear demand price regressions (Equation 4). OLS = ordinary least 

 squares; WLS = weighted least squares. 



'Computed from the formula (Afl^) (n-k-1)/(1 - H^) (r) where r, R^. and (n-k-1) represent the number 

 of restrictions, coefficient of determination, and degrees of freedom of the unrestricted model in hier- 

 archial order (1), (2), and (3) 



229 



