@RISK: Error Calculation in Distribution Fitting with (x,y) Pairs

@RISK computes RMS error by squaring differences between observed and predicted y-values, averaging, and taking the square root to rank distribution fits.

Jan 14, 2026
Knowledge

Title

Answer

Applies to: @RISK 5.x and Up

How @RISK calculates RMS error for fit ranking

Explanation
For curve data, @RISK measures goodness of fit by comparing the observed y-values (heights of probability density or relative frequency curves) with the fitted distribution’s predicted y-values. The RMS error is computed as the square root of the mean of the squared differences between the observed and predicted values.

RMS error = √[Σ(observed – predicted)² / n]
This calculation can be applied to normalized or unnormalized (x, y) data.
The resulting RMS error is used by @RISK to rank the quality of different fitted distributions.

Example
An attached workbook demonstrates how @RISK calculates RMS error using RiskFit functions (introduced in @RISK 6.0).

Attachments:
RMS_XY.xlsx

URL Name

rms-error-calculation-in-distribution-fitting-with-xy-pairs

@RISK: ​​​​​​​Error Calculation in Distribution Fitting with (x,y) Pairs

Information

Related Articles

Trending Articles

@RISK: Error Calculation in Distribution Fitting with (x,y) Pairs