Application of the Tweedie distribution to zero-catch data in CPUE analysis

Wefocuson the zero-catchproblemofCPUE (catchperunit effort) standardization. Because the traditional CPUE model with a log-normal error structure cannot be applied in this case, three methods have often been utilized as follows: (1) Ad hoc method adds a small constant value to all response variables. (2) Catch model with a Poisson or negative-binomial (NB) error structure. (3) Delta-type two-step method such as the delta-normal model (after estimating the ratio of zero-catch using a logit or probit model, a model such as CPUE log-normal or Catch-Poisson is applied to CPUE without Tweedie distribution model Zero-catch problem zero-data). e stat d out ch da ls) in , Thun atch i m th ibutio candi However, there are som In this paper, we carrie based on the actual by-cat by Japanese training vesse as a target (yellowfin tuna where the ratio of zero-c Poisson model derived fro of fish) has a gamma distr We also compared four (delta-type two-step method) terms of the statistical perform based on the observed CPUE an As a result, the differences in ad hoc method and Tweedie (target species). However, the s lognormal model, the Catch-NB species). Standardized CPUE ye the Tweedie distribution and o judging from the 5-fold cross-v by-catch species.