TY - BOOK ID - 136621833 TI - Machine Learning in Insurance AU - Nielsen, Jens Perch AU - Asimit, Alexandru AU - Kyriakou, Ioannis PY - 2020 PB - Basel, Switzerland MDPI - Multidisciplinary Digital Publishing Institute DB - UniCat KW - deposit insurance KW - implied volatility KW - static arbitrage KW - parameterization KW - machine learning KW - calibration KW - dichotomous response KW - predictive model KW - tree boosting KW - GLM KW - validation KW - generalised linear modelling KW - zero-inflated poisson model KW - telematics KW - benchmark KW - cross-validation KW - prediction KW - stock return volatility KW - long-term forecasts KW - overlapping returns KW - autocorrelation KW - chain ladder KW - Bornhuetter–Ferguson KW - maximum likelihood KW - exponential families KW - canonical parameters KW - prior knowledge KW - accelerated failure time model KW - chain-ladder method KW - local linear kernel estimation KW - non-life reserving KW - operational time KW - zero-inflation KW - overdispersion KW - automobile insurance KW - risk classification KW - risk selection KW - least-squares monte carlo method KW - proxy modeling KW - life insurance KW - Solvency II KW - claims prediction KW - export credit insurance KW - semiparametric modeling KW - VaR estimation KW - analyzing financial data KW - n/a KW - Bornhuetter-Ferguson UR - https://www.unicat.be/uniCat?func=search&query=sysid:136621833 AB - Machine learning is a relatively new field, without a unanimous definition. In many ways, actuaries have been machine learners. In both pricing and reserving, but also more recently in capital modelling, actuaries have combined statistical methodology with a deep understanding of the problem at hand and how any solution may affect the company and its customers. One aspect that has, perhaps, not been so well developed among actuaries is validation. Discussions among actuaries’ “preferred methods” were often without solid scientific arguments, including validation of the case at hand. Through this collection, we aim to promote a good practice of machine learning in insurance, considering the following three key issues: a) who is the client, or sponsor, or otherwise interested real-life target of the study? b) The reason for working with a particular data set and a clarification of the available extra knowledge, that we also call prior knowledge, besides the data set alone. c) A mathematical statistical argument for the validation procedure. ER -