On this page
Inspect the original 95% example
The CSV uses mean loss 0 and dispersion 10. It records the standard normal quantile, density at that quantile, selected 5% mass, tail-mean multiplier and both scaled outputs. All values are static teaching entries.
Reconcile the coefficient layers
The truncated upper-tail first moment equals φ(z), approximately 0.1031356404. Dividing by 0.05 gives the multiplier approximately 2.0627128075. Multiply by 10 and add zero to obtain ES approximately 20.6271280751.
Keep the threshold calculation distinct
VaR uses z≈1.644853627, not the tail-mean multiplier. With dispersion 10 its threshold is approximately 16.4485362695. The two coefficients answer different questions about the same normal law.
Update every linked value
If adapting the record, retain full precision in the calculation and round only the displayed results. Keep confidence and loss units explicit; do not rename a coefficient as an already scaled currency loss.