MMP.7:8 - Common Anti-Patterns and How to Avoid Them

Failure in this work	Repair
Reported-case frequency is substituted for population frequency despite selective reporting.	Derive the included-case law and the relation to the target rate.
An interval report is replaced by an event at its endpoint.	Sum or integrate the joint subject-and-recording law over the underlying values that can produce the report.
A common influence is independently removed from every observation.	Keep it in the joint law before elimination.
A likelihood is read as a probability distribution over the unknown parameter.	Supply the inferential method that turns it into the requested result.
A fitted model is trusted because its simulator reproduces its own assumptions.	Separate computational agreement from the subject comparison needed for the use.