Negative binomial models for development triangles of counts
By: Luis E. Nieto-Barajas, Rodrigo S. Targino
Potential Business Impact:
Counts future insurance claims more accurately.
Prediction of outstanding claims has been done via nonparametric models (chain ladder), semiparametric models (overdispersed poisson) or fully parametric models. In this paper, we propose models based on negative binomial distributions for the prediction of outstanding number of claims, which are particularly useful to account for overdispersion. We first assume independence of random variables and introduce appropriate notation. Later, we generalise the model to account for dependence across development years. In both cases, the marginal distributions are negative binomials. We study the properties of the models and carry out bayesian inference. We illustrate the performance of the models with simulated and real datasets.
Similar Papers
Flexible model for varying levels of zeros and outliers in count data
Methodology
Better counts for tricky data with many zeros.
Modeling Latent Underdispersion with Discrete Order Statistics
Methodology
Makes counting data more accurate for science.
Negative binomial regression and inference using a pre-trained transformer
Machine Learning (Stat)
Finds patterns in data much faster.