var (DataMatrix)

Return variance values in DataMatrix object

Syntax

```V = var(DMObj) V = var(DMObj, Flag) V = var(DMObj, Wgt) V = var(..., Dim) V = var(..., Dim, IgnoreNaN) ```

Input Arguments

 `DMObj` DataMatrix object, such as created by `DataMatrix` (object constructor). `Flag` Scalar specifying how to normalize the data. Choices are:`0` — Default. Normalizes using a sample size of N – `1`, unless N = `1`, in which case, normalizes using a sample size of `1`.`1` — Normalizes using a sample size of N.N = the number of elements in each column or row, as specified by `Dim`. For more information on the normalization equations, see the function `std`. `Wgt` Weight vector equal in length to the dimension over which `var` operates (specified by `Dim`. It is used to compute the variance. `Dim` Scalar specifying the dimension of `DMObj` to calculate the variances. Choices are:`1` — Default. Returns variance values for elements in each column.`2` — Returns variance values for elements in each row. `IgnoreNaN` Specifies if NaNs should be ignored. Choices are `true` (default) or `false`.

Output Arguments

 `V` An unbiased estimator of the variance within the columns or rows of a DataMatrix object. It can be either of the following:Row vector containing the variance values from elements in each column in `DMObj` (when `Dim` = `1`)Column vector containing the variance values from elements in each row in `DMObj` (when `Dim` = `2`)

Description

`V = var(DMObj)` returns the variance values of the elements in the columns of a DataMatrix object, treating NaNs as missing values. The data is normalized using a sample size of N`1`, where N = the number of elements in each column. `V` is a row vector containing the variance values for elements in each column in `DMObj`. The variance is the square of the standard deviation.

`V = var(DMObj, Flag)` specifies how to normalize the data. If `Flag` = `0`, normalizes using a sample size of N`1`. If `Flag` = `1`, normalizes using a sample size of N. N = the number of elements in each column or row, as specified by `Dim`. For more information on the normalization equations, see the function `std`. Default `Flag` = `0`.

`V = var(DMObj, Wgt)` computes the variance using `Wgt`, a weight vector whose length must equal the length of the dimension over which `var` operates (specified by `Dim`). All elements in `Wgt` must be nonnegative. The `var` function normalizes `Wgt` to sum of `1`.

`V = var(..., Dim)` returns the variance values of the elements in the columns or rows of a DataMatrix object, as specified by `Dim`. If `Dim` = `1`, returns `V`, a row vector containing the variance values for elements in each column in `DMObj`. If `Dim` = `2`, returns `V`, a column vector containing the variance values for elements in each row in `DMObj`. Default `Dim` = `1`.

`V = var(..., Dim, IgnoreNaN)` specifies if NaNs should be ignored. `IgnoreNaN` can be `true` (default) or `false`.