The Cake Guide to Cyber Risk Quantification: Understanding Lognormal Distributions for Absolute Beginners

Source: Medium

Author: Mehdi

URL: https://medium.com/@mpmab1/the-cake-guide-to-cyber-risk-quantification-understanding-lognormal-distributions-for-absolute-b31ee12daaa3

ONE SENTENCE SUMMARY:

This beginner’s guide explains lognormal distributions and their application in cyber risk quantification, using intuitive analogies and Python.

MAIN POINTS:

1. Lognormal distribution is essential in cyber risk quantification (CRQ).
2. Aimed at beginners without prior statistics knowledge.
3. Uses intuitive analogies like cake and cars.
4. Describes why averages are misleading in skewed data.
5. Explains transforming, validating, and reverse-transforming lognormal data.
6. Python and Monte Carlo simulations model cyber loss scenarios.
7. Visualize results with histograms and CDFs.
8. Lognormal properties: only positive, skewed, starts at zero, log is normal.
9. Misleading averages corrected by data transformation.
10. Applies to real-world scenarios like incomes and cyber losses.

TAKEAWAYS:

1. Averages in lognormal distributions can be misleading.
2. Log transformations stabilize data for better analysis.
3. Exponentiation returns data to its original scale.
4. Visualizing data helps to identify skewness and distribution types.
5. Monte Carlo simulations provide insights into possible outcomes.