Software for the evaluation of instrument calibration curves
Provides three regression techniques for fitting (fractional) polynomial curves:
- Ordinary least-squares regression (OLS)
- Weighted least-squares regression (WLS)
- Weighted total least-squares regression (WTLS)
This software is the main deliverable of task D1.3.5 Software for calibration problems developed and tested of the EMRP project Novel mathematical and statistical approaches to uncertainty evaluation (EMRP NEW04). Although the software was developed specifically for the calibration of flow meters, it can easily be applied to calibration data fitting for any other instrument type.
- Evaluates calibration curves fitting n pairs of measurement data (x, y)
- Uncertainties and correlations associated with both x and y values can be provided as input to the fitting procedures
- For each data point (x, y), single or repeated measurements can be provided as input to the fitting procedures
- Estimates of the fitting curve parameters and associated covariance matrix
- Normalized chi-squared value of the fit
- Fitted y values and associated covariance matrix
- Plot of the fitting curve with uncertainty bars on the y-axis
Release 1.3 (2015) was developed in MATLAB R2013a, with Optimization ToolboxTM 6.4, on a personal computer running Microsoft Windows 7.
- Works on 64-bit Microsoft Windows operating system.
- To install the software, extract the application installer CCC_Software_Installer_w64.exe from the archive CCC_Software_for_redistribution.zip and run it.
- Requires MATLAB's Compiler Runtime (MCR) version 8.3 (R2014a). The MCR installer file is not included in the CCC Software distribution, but it can be automatically downloaded from the internet during the installation process (see also README.txt).
- See the User Manual for a complete description of the software and for information on the installation.
- See the License Agreement for terms and conditions for the software license.
The authors are grateful to the National Physical Laboratory (NPL) colleagues for assisting with validating and testing Release 1.1 of the software, and for fruitful discussions on a preliminary version thereof.
The authors wish also to thank all the colleagues of the EMRP NEW04 Project for very useful comments and feedback on a previous Release 1.2 of the software.
The EMRP is jointly funded by the EMRP participating countries within EURAMET and the European Union.