Known Issue 139

Analyses using U.S. Severe Thunderstorm CLFs from Touchstone 8.0 or 9.0 in Touchstone Re 11.0 produce understated losses, and portfolio analyses may have unrepresentative correlations for U.S. Severe Thunderstorm.

Issue ID	139
Summary	Analyses using U.S. Severe Thunderstorm CLFs from Touchstone 8.0 or 9.0 in Touchstone Re 11.0 produce understated losses, and portfolio analyses may have unrepresentative correlations for U.S. Severe Thunderstorm.
Description	This impacts Model 20 CLFs of aggregate model version 1.3.0. The CLFs originate from Touchstone 8.0 or 9.0 when running analysis against the Model 22 version 8.1.0 ILF introduced in Touchstone Re 11.0. The analysis will complete without error; however, the Event IDs are unchanged. The Event IDs are only re-labeled as Model 22 instead of as Model 20. The results will match the original analysis used to generate the Model 20 CLF. The expected behavior is that the Event IDs would be updated and each event in the Verisk Severe Thunderstorm Model for the United States (M22) catalog would be represented by the closest match in the CLF. This is intended to reflect the increased event count in the latest Model 22 catalog. The greatest impact of this would be in analyzing a portfolio that has a mix of Model 20 v1.3.0 CLFs with newer M22 CLFs and/or Sums Insured programs. These combinations will result in spurious correlations in the U.S. Severe Thunderstorm results.
Resolved in	Touchstone Re 11.5
Date posted	8/30/2023
Software versions affected	11.0
Feature set	Aggregate Analysis
Models
Potential for loss numbers impact	Yes, especially for portfolios. See Detail.
Workaround	Analyzing the CLF using the World All Perils Archive event sets in Touchstone Re 11.0 produces the expected results for these CLFs containing Model 20 with proper event mapping. Furthermore, a DLF exported from such an analysis using a World All Perils Archive event set would then be correctly handled by the current World All Perils event set in Touchstone Re 11.0.