Proving prediction prudence
Author:
Open Access
- 1 January 2022
- journal article
- research article
- Published by American Institute of Mathematical Sciences (AIMS) in Data Science in Finance and Economics
- Vol. 2 (4) , 335-355
Abstract
We study how to perform tests on samples of pairs of observations and predictions in order to assess whether or not the predictions are prudent. Prudence requires that the mean of the difference of the observation-prediction pairs can be shown to be significantly negative. For safe conclusions,we suggest testing both unweighted (or equally weighted) and weighted means and explicitly taking into account the randomness of individual pairs. The test methods presented are mainly specified as bootstrap and normal approximation algorithms. The tests are general but can be applied in particular in the area of credit risk,both for regulatory and accounting purposes.Keywords
This publication has 11 references indexed in Scilit:
- Exposure at default modeling – A theoretical and empirical assessment of estimation approaches and parameter choiceJournal of Banking & Finance, 2018
- The Validation of Risk Models2016
- Objective Bayesian Estimation of the Probability of DefaultJournal of the Royal Statistical Society Series C: Applied Statistics, 2015
- A proposed framework for backtesting loss given default modelsJournal of Risk Model Validation, 2014
- A statistical repertoire for quantitative loss given default validation: overview, illustration, pitfalls and extensionsJournal of Risk Model Validation, 2014
- The art of probability-of-default curve calibrationJournal of Credit Risk, 2013
- Validation techniques and performance metrics for loss given default modelsJournal of Risk Model Validation, 2009
- Modern Applied Statistics with S2002
- Stochastic Processes for Insurance & Finance1999
- Bootstrap Methods and their Application1997