ABSTRACT
This chapter discusses how to fit measurement data to a model using the least squares method, and also discusses error and accuracy in data fitting. For a model to make sense, the data points should be scattered randomly around the model predictions: the mean of the error should be zero. It should be checked how the error evolves and if there is or is not a correlation with the measured value. If the former is the case, it probably indicates that something is wrong with the model structure. Analysis of variance (ANOVA) will calculate F-values on the basis of the degrees of freedom and the sum of squared errors to see whether certain factors are significant or not. The chapter also postulates expressions for the confidence limits for the fit parameters and the predicted points.
