Sample Standard Deviation - Combining Standard Deviations - Sample-based Statistics

Sample-based Statistics

Standard deviations of non-overlapping (X ∩ Y = ∅) sub-samples can be aggregated as follows if the actual size and means of each are known:

\begin{align} \mu_{X \cup Y} &= \frac{1}{N_{X \cup Y}}\left(N_X\mu_X + N_Y\mu_Y\right)\\ \sigma_{X \cup Y} &= \sqrt{\frac{1}{N_{X \cup Y} - 1}\left(\sigma_X^2 + N_X\mu_X^2 + \sigma_Y^2 + N_Y\mu _Y^2 - \mu_{X \cup Y}^2\right) } \end{align}

For the more general case of M non-overlapping data sets, X1 through XM, and the aggregate data set :

\begin{align} \mu_X &= \frac{1}{\sum_i { N_{X_i}}} \left(\sum_i { N_{X_i} \mu_{X_i}}\right)\\ \sigma_X &= \sqrt{\frac{1}{\sum_i {N_{X_i} - 1}} \left( \sum_i { \left } - \left\mu_X^2 \right) } \end{align}

where:

If the size, mean, and standard deviation of two overlapping samples are known for the samples as well as their intersection, then the standard deviation of the aggregated sample can still be calculated. In general:

\begin{align} \mu_{X \cup Y} &= \frac{1}{N_{X \cup Y}}\left(N_X\mu_X + N_Y\mu_Y - N_{X\cap Y}\mu_{X\cap Y}\right)\\ \sigma_{X \cup Y} &= \sqrt{ \frac{1}{N_{X \cup Y} - 1}\left(\sigma_X^2 + N_X\mu_X^2 + \sigma_Y^2 + N_Y\mu _Y^2 - \sigma_{X \cap Y}^2 - N_{X \cap Y}\mu_{X \cap Y}^2 - \mu_{X \cup Y}^2\right) } \end{align}

Read more about this topic:  Sample Standard Deviation, Combining Standard Deviations

Famous quotes containing the word statistics:

    “Maybe a nation that consumes as much booze and dope as we do and has our kind of divorce statistics should pipe down about “character issues.” Either that or just go ahead and determine the presidency with three-legged races and pie-eating contests. It would make better TV.”
    —P.J. (Patrick Jake)