crab-project | commercial catch of dungeness crab

crab-project is a Python library. crab-project has no bugs, it has no vulnerabilities and it has low support. However crab-project build file is not available. You can download it from GitHub.

the goal of my project was to predict the commercial catch (aka 'landings') of dungeness crab in the northern california eureka area. the eureka area (defined by port from fort bragg northward to the oregon border) historically has contributed about 3/4 of all dungeness crab landings for the state, but has also experienced dramatic and unpredictable year-over-year fluctuations. to address this uncertainty, i created a predictive model that estimates landings in this crucial region before the season opens. the best-performing model incorporates data of landings from previous seasons as well as relevant exogenous environmental factors sampled from 3-4 years prior to the target prediction year. these findings are reasonable in the biological sense in that crabs of this type are typically caught at age 3-4. my results were encouraging, as the best model shows not only increased predictive power over alternative time series models, but is also consistently robust across the testing window of the past 50 years. comparing the model to a naive model such as the 'rolling mean' predictor was essential in justifying that a significant findings were found. using leave-one-out cross validation across the entirety of the testing window, the mae of the residuals was decreased from