Window

Rolling objects are returned by .rolling calls: pandas.DataFrame.rolling(), pandas.Series.rolling(), etc. Expanding objects are returned by .expanding calls: pandas.DataFrame.expanding(), pandas.Series.expanding(), etc. EWM objects are returned by .ewm calls: pandas.DataFrame.ewm(), pandas.Series.ewm(), etc.

Standard moving window functions

Rolling.count() rolling count of number of non-NaN
Rolling.sum(*args, **kwargs) rolling sum
Rolling.mean(*args, **kwargs) rolling mean
Rolling.median(**kwargs) rolling median
Rolling.var([ddof]) rolling variance
Rolling.std([ddof]) rolling standard deviation
Rolling.min(*args, **kwargs) rolling minimum
Rolling.max(*args, **kwargs) rolling maximum
Rolling.corr([other, pairwise]) rolling sample correlation
Rolling.cov([other, pairwise, ddof]) rolling sample covariance
Rolling.skew(**kwargs) Unbiased rolling skewness
Rolling.kurt(**kwargs) Unbiased rolling kurtosis
Rolling.apply(func[, args, kwargs]) rolling function apply
Rolling.quantile(quantile, **kwargs) rolling quantile
Window.mean(*args, **kwargs) window mean
Window.sum(*args, **kwargs) window sum

Standard expanding window functions

Expanding.count(**kwargs) expanding count of number of non-NaN
Expanding.sum(*args, **kwargs) expanding sum
Expanding.mean(*args, **kwargs) expanding mean
Expanding.median(**kwargs) expanding median
Expanding.var([ddof]) expanding variance
Expanding.std([ddof]) expanding standard deviation
Expanding.min(*args, **kwargs) expanding minimum
Expanding.max(*args, **kwargs) expanding maximum
Expanding.corr([other, pairwise]) expanding sample correlation
Expanding.cov([other, pairwise, ddof]) expanding sample covariance
Expanding.skew(**kwargs) Unbiased expanding skewness
Expanding.kurt(**kwargs) Unbiased expanding kurtosis
Expanding.apply(func[, args, kwargs]) expanding function apply
Expanding.quantile(quantile, **kwargs) expanding quantile

Exponentially-weighted moving window functions

EWM.mean(*args, **kwargs) exponential weighted moving average
EWM.std([bias]) exponential weighted moving stddev
EWM.var([bias]) exponential weighted moving variance
EWM.corr([other, pairwise]) exponential weighted sample correlation
EWM.cov([other, pairwise, bias]) exponential weighted sample covariance