Variance
When computing measures to describe the dispersion or variability of a distribution, we may look at 'spread' or 'deviation' of values from the mean value. We can then represent this 'spread' by reporting an index like 'mean deviation from the mean'. However, sum of deviation about the mean is 0 and thus 'mean deviation from the mean' turns out to be 0, irrespective of the spread in the distribution. This can be taken care of by squaring the deviation values and then summating them. Thus, the variance of N observations is the 'mean of squared deviations' and is
Algebraically, variance is square of standard deviation, s2. However, standard deviation is used instead of variance because the units of the calculated variance don't make sense as they are to the power of 2 and not in the same units as the data itself.
See Also:
standard deviation, coefficient of variation
References:
Frank, H. and Althoen, S.C. "The variance and standard deviation." §C.4 in Statistics: Concepts and Applications Cambridge, Great Britain: Cambridge University Press, pp. 52-57, 1995.
http://mathworld.wolfram.com/Variance.html
Cite This As:
Dogra, Shaillay K., "Variance." From QSARWorld--A Strand Life Sciences Web Resource.
http://www.qsarworld.com/qsar-statistics-variance.php
|
