Safe iterative Bayesian model building

Description of the granted funding

Statistical analysis is critical when it comes to obtaining insights from data. Despite the practical success of iterative Bayesian statistical model building, it has been criticized to violate pure Bayesian theory and that we may end up with a different model had the data come out differently. In this project, we formalize and develop theory and diagnostics for safe iterative Bayesian model building. We show that when the iterative model building is done carefully, the difference to the theoretically optimal result is negligible. The practical diagnostics guide the modeller through the appropriate steps to ensure safe iterative model building, or indicate when the modeller is likely to be in the danger zone. By making applied scientific research and data analysis in private and public sectors more reliable and reproducible, our understanding of the world and decision-making will be improved. Thus, the project will have a long lasting positive impact on society.
Show more

Starting year

2021

End year

2025

Granted funding

Aki Vehtari Orcid -palvelun logo
531 066 €

Funder

Research Council of Finland

Funding instrument

Academy projects

Other information

Funding decision number

340721

Fields of science

Computer and information sciences

Research fields

Laskennallinen data-analyysi

Identified topics

computer science, information science, algorithms