|Error Analysis and Data Analysis by Least Square Fitting|
For a small set of data, the best estimate of the standard deviation is given by
Maximum Possible Error:
In the above discussion we have only considered a measurement of one quantity. Most of the experiments involve measurement of several different quantities. Measurement of each of these quantities is limited in accuracy by the least count of the instrument and error arising because of this is termed as maximum possible error. It can be estimated in the following manner .
Suppose a physical quantity y is a function of two independent ( measurable) variables x1 and x2 given by,