Bayesian Calibration
Infer parameter distributions rather than one best-fit point.
Build the mental model first.
Calibration uncertainty is information; retain it instead of hiding it behind one optimiser output.
Infer parameter distributions rather than one best-fit point.
Now make it exact.
Notation and units
Decimal rates and volatilities, year-fraction time and continuous compounding unless stated otherwise.
Every model has a price.
Educational conventions are stated explicitly and may simplify market quotation or settlement details.
Rates are continuously compounded unless the section says otherwise.
Inputs are deterministic in the base model.
“An unstated convention is a future reconciliation break.”— THEQUANTBATEMAN
Why a quant cares.
Research and specialist risk analysis, especially where parameter uncertainty matters.
The hedge has opinions.
Start with the quote convention, then ask which Frontier risk survives the hedge. A number without its convention is merely well-dressed ambiguity.
Ask Bateman about this model →