A Note on the Application of the Normal Power Method When Estimating Probable Yearly Aggregate
这篇注记说明了当使用季度或月度损失数据而非年度数据时，应用正态幂方法估计最大可能年度总损失所需做出的修改，对保险业的风险评估有参考价值。
In the important work by Cummins and Freifelder [1], the normal power method is shown to be an excellent statistical tool for estimating the maximum probable yearly aggregate loss (MPY)' for many of the skewed distributions found in property/liability insurance claims processes. In attempting to apply this methodology to actual loss data, however, it is often necessary to use quarterly or monthly loss figures2 in order to assemble a sufficient number of data points to provide credible analysis.3 The purpose of this note is to clearly show the modifications that must be made when applying the normal power method to data points more frequent than annual.