Prediction of measurement standard values to improve uncertainty, under well-defined drift conditions
The behaviour of the values of some measurement standards, in many cases, have well-defined drifts and therefore can be modelled by linear regressions, using the least squares method. The linear regression model allows to predict the measurement standard values with their respective uncertainties, w...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-12-01
|
| Series: | Measurement: Sensors |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2665917421000453 |
| Summary: | The behaviour of the values of some measurement standards, in many cases, have well-defined drifts and therefore can be modelled by linear regressions, using the least squares method. The linear regression model allows to predict the measurement standard values with their respective uncertainties, which can be better than the measurement uncertainties. The regression method used is generalized least squares (GLS) and the uncertainty of the prediction values is estimated by Monte Carlo simulation. |
|---|---|
| ISSN: | 2665-9174 |