Regresión cuantílica como punto de partida en los modelos predictivos para el riesgo

Albert Pitarque; Ana María Pérez-Marín; Montserrat Guillen

doi:10.26360/2019_5

Authors

Albert Pitarque Universidad de Barcelona (España)
Ana María Pérez-Marín Universidad de Barcelona (España)
Montserrat Guillen Universidad de Barcelona (España)

DOI:

https://doi.org/10.26360/2019_5

Keywords:

predictive modelling, value-at-risk, tail value at-risk, optimization, resampling

Abstract

Given a risk level or tolerance, quantile regression is a predictive model that fits the corresponding percentile of the continuous response variable. Given a fixed percentage value, we identify the effect of each predictor variable in the cumulative distribution up to that level of the dependent variable. In this article, we show how this methodology can be used in motor insurance data analysis and we propose an extension of quantile regression inspired by the need to predict the expectation of the conditional tail. To this end, specific R routines have been developed and a resampling procedure has been implemented to approximate standard errors. The main conclusion is that this type of models allows us to analyze which factors affect accident risk and can be used to mitigate or to evaluate risk in the insurance field.

Downloads

Download data is not yet available.

References

Acerbi, C., y Szekely, B. (2014). Back-testing expected shortfall. Risk, 27(11), 76-81.

Ayuso, M., Guillen, M., y Pérez-Marín, A. M. (2016a). Using GPS data to analyse the distance travelled to the first accident at fault in pay-as-you-drive insurance. Transportation Research Part C: Emerging Technologies, 68, 160-167.

Ayuso, M., Guillen, M., y Pérez-Marín, A. M. (2016b).Telematics and gender discrimination: some usage-based evidence on whether men’s risk of accidents differs from women’s. Risks, 4(2), 1-10.

Boucher, J-P., Côté, S., y Guillen, M. (2017). Exposure as duration and distance in telematics motor insurance using generalized additive models. Risks, 5(4), 54. https://doi.org/10.3390/risks5040054

Fissler, T., y Ziegel, J. F. (2016). Higher order elicitability and Osband's principle. The Annals of Statistics, 44(4), 1680-1707.

Guillen, M., Nielsen, J. P., Ayuso, M., y Pérez-Marín, A. M. (2019). The use of telematics devices to improve automobile insurance rates. Risk Analysis, 39(3), 662-672.

Hardy, M. R. (2006). An introduction to risk measures for actuarial applications. SOA Syllabus Study Note. Koenker, R., y Bassett, G. (1978). Regression quantiles. Econometrica, 46(1), 33-50.

Koenker, R., y Machado, A. F. (1999). Goodness of fit and related inference processes for quantile regression. Journal of the American Statistical Association, 94(448), 1296-1310.

Koenker, R., Portnoy, S., Ng, P. T., Zeileis, A., Grosjean, P., y Ripley, B. D. (2018). Package 'quantreg'. Cran R-project.

Pitarque, A. (2019). Regresió quantílica en la gestió de riscos. Trabajo de final de máster Universitat Politècnica de Catalunya y Universitat de Barcelona.

Pitt, D. G. W. (2006). Regression quantile analysis of claim termination rates for income protection insurance. Annals of Actuarial Science, 1(2), 345- 357.