operator

class timeseries.operator._TimeSeriesOperator(tseries)

Bases: object

Interface class for extending set operations to time series based on dates and applying functions to resulting time series.

Parameters

tseries – time series to operate on

_apply(func, arg, operation, elementwise, fill, **kwargs)

Apply function to time series after set operation.

Parameters

• func – function to apply to time series

• arg – time series or broadcasted argument passed to function

• operation – set operation to apply to time series dates

• elementwise – flag if function is applied elementwise and should return time series

• fill – value to fill in missing dates

• **kwargs – keyword arguments passed to function

classmethod _apply_function(tseries, arg, func, elementwise, **kwargs)

Apply function to time series and return result, or results as timeseries if elementwise.

Parameters

• tseries – time series to operate on

• arg – time series or broadcasted argument passed to function

• func – function to apply to time series

• elementwise – flag if function is applied elementwise and should return time series

• **kwargs – keyword arguments passed to function

static _broadcast(value, size)

Return value broadcasted to tuple.

Parameters

• value – float-convertable value

• size – size of tuple

classmethod _fill_tseries(dates, tseries, fill)

Return time series in specified dates with missing values filled.

Parameters

• tseries – time series object to extend or contract

• dates – dates to keep and fill if missing

• fill – value to fill in missing dates

static set_to_dates(set)

Convert set of dates to chronologically sorted tuples.

custom(func, arg, operation='intersection', elementwise=True, fill=0, **kwargs)

Apply custom function to time series after set operation.

Parameters

• func – function to apply to time series

• arg – time series or broadcasted argument passed to function

• operation – set operation to apply to time series dates, defaults to ‘intersection’

• elementwise – flag if function is applied elementwise and should return time series, defaults to True

• fill – value to fill in missing dates, defaults to 0

• **kwargs – keyword arguments passed to function

timeseries.operator.add(values1, values2)

Return the element-wise sum of two numeric iterables.

Parameters

• values1 – finite-length iterable of float-convertable numbers

• values2 – finite-length iterable of float-convertable numbers

timeseries.operator.subtract(values1, values2)

Return element-wise difference of two numeric iterables.

Parameters

• values1 – finite-length iterable of float-convertable numbers

• values2 – finite-length iterable of float-convertable numbers to subtract

timeseries.operator.multiply(values1, values2)

Return element-wise multiplication of two numeric iterables.

Parameters

• values1 – finite-length iterable of float-convertable numbers

• values2 – finite-length iterable of float-convertable numbers

timeseries.operator.divide(values1, values2)

Return element-wise division of two numeric iterables.

Parameters

• values1 – finite-length iterable of float-convertable numbers

• values2 – finite-length iterable of float-convertable numbers to divide with

timeseries.operator.right_divide(values1, values2)

Return element-wise right division of two numeric iterables.

Parameters

• values1 – finite-length iterable of float-convertable numbers to divide with

• values2 – finite-length iterable of float-convertable numbers

timeseries.operator.power(values1, values2)

Return element-wise exponentiation of two numeric iterables.

Parameters

• values1 – finite-length iterable of float-convertable numbers

• values2 – finite-length iterable of float-convertable numbers to use as exponent

timeseries.operator.right_power(values1, values2)

Return element-wise right exponentiation of two numeric iterables.

Parameters

• values1 – finite-length iterable of float-convertable numbers to use as exponent

• values2 – finite-length iterable of float-convertable numbers