Journal article
North American Actuarial Journal · 2019-10-02 · Volume 23 · Issue 4 · pp. 551-572
DOI: 10.1080/10920277.2019.1634595
Published online: 2019-10-18 · Download citation (BibTeX)
Abstract
A new relational data-matching model is presented to predict individual farmer yields in the absence of farm-level data. The relational model defines a similarity measure based on an Euclidean distance metric that considers weather information, farm size, county size and the coefficient of variation of yield, to search for the most "similar" region in a different country to borrow individual loss experience data that is otherwise not available. Detailed farm-level and county-level corn yield data in Canada and the U.S. are used to empirically evaluate the proposed relational model. The results show that the model achieves lower mean and standard deviation prediction errors and recovers the actual premium rate more accurately compared to the benchmark model. This research provides a new approach for insurers, reinsurers and governments to enhance individual loss experience, helping to overcome issues such as data scarcity and credibility, as well as aggregation bias, that present substantial challenges in risk modelling, pricing and developing new insurance programs, particularly in developing countries.
Keywords: Relational Model, Aggregation Bias, Shortness of Data, Euclidean Distance, Crop Insurance, Yield Forecasting, Ratemaking