Rolling Origin Forecast Validation Calculator
Evaluate one-step trailing-mean forecasts using only values observed before each held-out time.
YOUR CALCULATION
Result
—
Calculation steps
Work through one step at a time or reveal the complete calculation. The steps use your current inputs.
How to calculate
Expand the available training prefix one observation at a time without using future values.
At time t, predict mean of previous min(window,t) values; compare with held-out yt
Worked example
For10,12,14,16,18,20 with initial3 and window2, forecasts13,15,17 have errors−3,−3,−3.
Assumptions and input details
One-step trailing mean only, equal observation intervals. No tuning on these same scores, seasonal model, multi-step horizon or future performance guarantee.
Using this calculator
Can I enter my own problem?
Yes. Replace the example values with your own inputs. The result, calculation steps, tables and applicable diagrams update from those inputs. Blank or invalid values clear the result; example values are never silently substituted for missing inputs.
Are the results exact?
Results use floating-point arithmetic and displayed rounding. Read the formula, assumptions and any stated data precision before interpreting the result.
Is my input uploaded?
The calculation runs in your browser. No account, file upload or external AI service is required. Inputs are not saved between page reloads.
Compare trailing simple moving averages with first-value-seeded exponential smoothing for an ordered series.
Evaluate paired actual and forecast values with signed bias, absolute errors and explicitly defined percentage metrics.
Calculate additive level/trend updates and ordinary or damped Holt point forecasts with explicit initialization and entered smoothing parameters.
Reserved advertising space.