Simulation Study Instructions

The goal of the simulation study is to implement your model(s) and evaluate their simulations to assess the differences between them. Your goal is not to “test” relevant hypotheses yet, but to visualize and describe the differences in the model hindcasts and projections.

Simulation Study Guidelines

The proposal should have the following components with 1 inch margins (on all sides) and at least 11 point font.

• Calibration: Find and report appropriate parameter values for your model(s). This can be done through finding the maximum likelihood or maximum a posteriori estimates, with relevant uncertainty estimates obtained from the bootstrap or sampling from the posterior distribution using MCMC. Whatever approach you take, make sure to justify it and any associated assumptions (e.g. prior distributions).
• Simulation: Using the calibrated model(s), simulate synthetic datasets and compute useful summaries. Include useful graphical checks (including, if appropriate, hindcasts and projections) as needed.
• Comparison: Discuss the differences in the calibrations and simulations and what they mean for your hypotheses(s). Were there any surprises?
• References: Make sure to include any references cited in your proposal.