Second Workshop on Estimation with the RDBES data model (WKRDB-EST2; outputs from 2020 meeting)

The Second Workshop on Estimation with the RDBES data model (WKRDB-EST2) met online in September 2020 to develop and document R scripts for design-based estimation in the RDBES data model, identify issues in the data model that impact that type of estimation and develop a roadmap for further development of estimation within the RDBES. The main outcomes of WKRDB-EST2 were:

• Discussion and test of a collaborative process, involving all stages of development (from function scripting to package maintenance);

• Collaborative development of data preparation and estimation functions in open-source code (GitHub);

• Carrying out of initial estimation tests that indicated data model suitable for design-based estimation;

• Progress in variance estimation to allow confidence intervals around estimates to be delivered;

• Kick-off of a package “icesRDBES” that will contain the functions, document them, quality check them and make them available to the wider community;

• Discussion of existing data model issues related to estimation and proposal of solutions.

In the final consultation among participants of this WK it was suggested that estimation work in RDBES should be set as a three years’ ICES fixed-term working group (WGRDBES-EST). Such WG was considered necessary to secure the intersessional engagement of the participants and a steady and continuous development of all the main estimators relevant for the ICES community including e.g. ratio estimators and procedures for handling of industry refusals. In addition to estimation, the WGRDBES-EST should also reflect on issues such as the long-term maintenance of the code it develops and on a system for peer-review of its work.