Examples
Examples
Try the starting example
- Dated errors JSON [elapsed days,error measurement units]
- [[0,0],[10,0.1],[20,0.2]]
- Permitted lower error (measurement units)
- -0.5
- Permitted upper error (measurement units)
- 0.5
- Observed OLS drift slope (units/day)
- 0.01
- Fitted day-zero error (measurement units)
- 0
- Fitted error at last observation (measurement units)
- 0.2
- Projected directional limit-crossing day
- 50
Try another scenario
- Dated errors JSON [elapsed days,error measurement units]
- [[0,0],[10,0.2],[20,0.4]]
- Permitted lower error (measurement units)
- -0.5
- Permitted upper error (measurement units)
- 0.5
- Observed OLS drift slope (units/day)
- 0.02
- Fitted day-zero error (measurement units)
- 0
- Fitted error at last observation (measurement units)
- 0.4
- Projected directional limit-crossing day
- 25
Calculation method
Calculation method
Observed OLS drift slope (units/day)
Cov(day,error)/sum squared centered days.
Fitted day-zero error (measurement units)
mean(error)-slope*mean(day).
Fitted error at last observation (measurement units)
intercept+slope*maxDay.
Projected directional limit-crossing day
Solve (directionalLimit-intercept)/slope; zero slope or past crossing returns explicit status.
Symbols and units
- a
- Dated errors JSON [elapsed days,error measurement units]
- b
- Permitted lower error (measurement units)
- c
- Permitted upper error (measurement units)
Use the units shown. A list keeps the order in which you enter its values. The symbol ^ means “raised to the power of”.
Assumptions & limits
Assumptions & limits
- Unweighted linear trend through dated errors. Projection is conditional, not a recommended calibration interval; no uncertainty, nonlinearity or manufacturer interval inferred. Supported limits: 2..1000 unique increasing day offsets>=0; finite errors; lower<upper; no constant-time series.
Are these inputs private?
These calculations run in your browser. A shared link contains the values you choose to share. Only questions submitted to the separate AI finder are sent to OpenAI.
Further reading
Background reading; this does not imply endorsement or independent review of SumFizz.