A variance function describes the relationship between the variance and the measured quantity value.
The are numerous models that describe the relationship between the variance and measured quantity value across the measuring interval:
||Fit constant variance across the measuring interval.
||Fit constant coefficient of variation across the measuring interval.
|Mixed constant / proportional variance
||Fit constant variance at low levels with constant coefficient of variation at high levels.
||Fir a 2-parameter linear variance fuction.
||Fit Sadler 3-parameter variance function – a monotone relationship (either increasing or decreasing) between the variance and level of measurement.
||Fit Sadler alternate 3-parameter variance function – gives more flexibility than the Sadler standard 3-parameter variance function especially near zero.
||Fit a Sadler 4-parameter variance function - allows for a turning point near the detection limit.
A variance function can be useful to estimate the limit of detection or limit of quantitation.