Author: Sam L. Savage
Source: Kozlova, M., & Yeomans, J. S. (Eds.). (2024). Sensitivity Analysis for Business, Technology, and Policymaking: Made Easy with Simulation Decomposition (SimDec). Taylor & Francis. https://doi.org/10.4324/9781003453789
License: CC BY-NC-ND 4.0
📖 Read full Afterword: 004_Afterword.pdf
🎬 Watch Sam presenting it: Youtube
Sam L. Savage connects the SimDec method with the field of probability management, particularly with Stochastic Information Packets (SIPs) and the SIPmath™ standard. These concepts allow uncertainty to be stored as data, enabling downstream calculations that preserve probability distributions.
SimDec, according to Savage, is a natural consumer of SIP libraries, allowing users to explore the “personalities” behind histograms — the contributing input combinations that shape uncertainty.
Savage humorously refers to decomposed histograms as "Schizograms" — revealing the internal logic of uncertain outcomes. For example, when SimDec is applied to Net Present Value (NPV) outputs across different combinations of price and demand, it visually uncovers distinct distributions. Each partition reveals how different factors interact and contribute to the overall shape of results.
He likens this process to placing a histogram on a "psychiatrist’s couch" — exploring its moods and modes rather than just accepting a single average or variance.
SIPmath was originally developed to aggregate complex simulations (e.g. oil and gas portfolios) into usable data across Excel, Python, and R. With vectorized calculations and tools like ChanceCalc, SIPmath now supports interactive simulations without macros — and can integrate SimDec-style decomposition visually.
When SimDec includes partition columns, tools like ChanceCalc can even recreate color-coded decompositions (Schizograms) directly in Excel.
Savage shares a personal example from a Royal Dutch Shell project, analyzing energy portfolios vulnerable to political risk. He simulated various gas projects, such as:
- African sources with high returns but political instability
- Scandinavian sources with low returns but stable supply
Using SimDec, he decomposed the portfolio SIP into contributions from each project, revealing how hedging Africa with Scandinavia remained optimal across a wide range of disruption probabilities (3% to 30%). This clarified the robustness of the hedge in a way that histograms alone could not.
The afterword concludes with a hopeful message: SimDec and SIPmath together can drive better modeling by:
- Embedding uncertainty directly into data
- Allowing full transparency of assumptions
- Making models interactive, visual, and explainable
SimDec becomes a visual interpreter for SIPmath data, enabling a new generation of model clarity.
Based on the Afterword by Sam L. Savage in Sensitivity Analysis for Business, Technology, and Policymaking
© Sam L. Savage, 2024 — CC BY-NC-ND 4.0
This summary is an independent derivative created for educational and indexing purposes, not affiliated with the original publisher.