Table 2. — Regression results using various weights for 

 tuna species holding independent variables constant. 



efficients of determination in the weighted 

 total pounds equation and the total pounds 

 equation is not statistically significant (0.70 

 vs. 0.68). 



The total pounds variable has, of course, 

 almost the same weights (Vs, Vs, V3) as the 

 weighted output variable so that, ultimately, 

 it may be of marginal significance to distinguish 

 between them in this fishery. Nevertheless, 

 we cannot know this before further experiments 

 are conducted. 



Total value as a dependent variable is inferior 

 to total pounds. This tends to confirm our 

 hypothesis that yellowfin and skipjack are joint 

 products in this fishery. The weight of skipjack 

 in total value is less than the weight for 

 yellowfin and bluefin.^ Therefore, it appears 

 that the amount for which skipjack can be sold 

 is not reflected in the extra work done in 

 catching it, at least relative to yellowfin and 

 bluefin. 



The best production functions for the tuna 

 fishery are shown in Table 3. The only fishing 

 time variable available for this fishery was 



'The "Best" solution. 



Source: Economic Research Laboratory, National Marine 

 Fisheries Service, 1970. 



■• The relative price weights are .286 for skipjack, 

 .376 for yellowfin, and .344 for bluefin. 



Table 3. — Tuna purse seine production function: alternate specifications. 



Dependent variable 



INDEPENDENT VARIABLE 

 LOG CAPACITY LOG DAYS LOGH.P.' 66 DUM 67 DUM 



yiNT. /?■ 



Problem 1 



Weighted total pounds 



Reg. Coef. .520 



/ratio 8.41 



Part Cor. Coef. .462 



.416 



5.12 

 .302 



.328 

 4.77 

 .283 



-.196 



.587 



76.17 



.453 



.680 



113.84 



.168 



.704 



127.07 



' Horsepower 



Source: Economic Research Laboratory, National Marine Fisheries Service, 1970. 



50 



