Relying on historic analysis when assessing potential risks and possible impacts implies that

Relying on historic analysis for risk assessment is predicated on the principle of stationarity—the assumption that the underlying conditions and processes that generated past risks will remain stable. This implies a core belief that past events, their frequencies, and their impacts are a reasonable proxy for what will happen in the future. While ISO 31000 acknowledges historical data as a valid input for risk analysis, it also stresses the importance of considering the limitations and confidence in that information, as this core assumption may not hold true in dynamic environments.

A. Historical models cannot guarantee accuracy for future events, as unforeseen factors or systemic changes can render past data irrelevant for predicting the impact of new adverse events.

B. This method inherently cannot account for novel or emerging risks ("black swans") that have no historical precedent, meaning not all significant risks can be confidently analyzed.

1. ISO 31000:2018, Risk management — Guidelines. Clause 6.4.3, "Risk analysis," states that the analysis should be based on the "best available information" and that "assumptions and limitations" of this information should be considered. This directly addresses the fact that relying on historical data is an assumption about its continued relevance to the future.

2. Aven, T. (2011). Quantitative Risk Assessment: The Scientific Platform. Cambridge University Press. In Chapter 2, Section 2.3, Aven discusses that probability-based risk assessments using historical data are founded on the assumption of a stable system where past frequencies are representative of future likelihoods.

3. MIT OpenCourseWare. (2016). 15.450 Analytics of Finance, Lecture 1: Introduction to Quantitative Finance. The course materials discuss the use of historical data for financial modeling, explicitly stating the key assumption is "stationarity," meaning the statistical properties (like mean and variance) of the data are constant over time. This is analogous to believing the future will behave like the past.