## Documentation Center |

Variance ignoring `NaN`s

y = nanvar(X) y = nanvar(X,1) y = nanvar(X,W) y = nanvar(X,W,DIM)

| Financial times series object. |

| Weight vector. |

| Dimension along which the operation is conducted. |

`nanvar` for financial times series objects
is based on the Statistics Toolbox™ function `nanvar`.
See `nanvar` in the Statistics Toolbox documentation.

`y = nanvar(X)` returns the sample variance
of the values in a financial time series object `X`,
treating `NaN`s as missing values. `y` is
the variance of the non-`NaN` elements of each series
in `X`.

`nanvar` normalizes `y` by `N` – `1` if `N` > `1`,
where `N` is the sample size of the non-`NaN` elements.
This is an unbiased estimator of the variance of the population from
which `X` is drawn, as long as `X` consists
of independent, identically distributed samples, and data are missing
at random. For `N` = `1`, `y` is
normalized by `N`.

`y = nanvar(X,1)` normalizes by `N` and
produces the second moment of the sample about its mean. `nanvar(X,
0)` is the same as `nanvar(X)`.

`y = nanvar(X,W)` computes the variance using
the weight vector `W`. The length of `W` must
equal the length of the dimension over which `nanvar` operates,
and its non-`NaN` elements must be nonnegative.
Elements of `X` corresponding to `NaN` elements
of `W`are ignored.

`y = nanvar(X,W,DIM)` takes the variance along
dimension `DIM` of `X`.

To compute `nanvar`:

f = fints((today:today+1)', [4 -2 1; 9 5 7]) f.series1(1) = nan; f.series3(2) = nan; nvar = nanvar(f)

nvar = 0 24.5000 0

Was this topic helpful?