Usage

To use finance_calculator in a project:

>>> import finance_calculator as fc

>>> drawdown = fc.get_drawdown(scheme_data)
>>> volatility = fc.get_volatility(scheme_data)
>>> sharpe = fc.get_sharpe(scheme_data, benchmark_data)
>>> sortino = fc.get_sortino(scheme_data, benchmark_data)
>>> treynor = fc.get_treynor(scheme_data, benchmark_data)
>>> alpha = fc.get_alpha(scheme_data, benchmark_data)
>>> beta = fc.get_beta(scheme_data, benchmark_data)
>>> upside_capture = fc.get_upside_capture(scheme_data, benchmark_data)
>>> downside_capture = fc.get_downside_capture(scheme_data, benchmark_data)

If you want only current value of a given ratio, you can use tail=True as a keyword argument in all of these functions. With tail=False it will give a pandas dataframe with values in a rolling window fashion.

By default it is assumed that data is for 250 days of a year. Also rolling window is taken to be 3 years by default. It can be used as fc.get_sharpe(scheme_data, benchmark_data, window=500, annualiser=250) for 2 years rolling data. Annualiser is meant for annualising the values. If your data contains 365 days in a year you should pass annualiser=365. Normally, if it contains only working days, it will by default take annualiser=250.

Instead of benchmark_nav_data, an annual value of risk_free_rate can be given. The calculator can use it in place of benchmark_nav_data. Internally, it will create a benchmark nav df from the risk_free_rate. The value can passed as fc.get_sharpe(scheme_data, risk_free_rate=0.05) for 5% of annual risk free returns.

Also, if multiple out of these are needed for one set of data only, you can get a ratio calculator instance and call the functions on that:

>>> import finance_calculator as fc
>>> rc = fc.get_ratio_calculator(nav_data, benchmark_nav_data)
>>> beta_df = rc.get_beta(window=250*3)
>>> alpha_df = rc.get_alpha(window=250*3)
# similarly other ratios can be called
# window needs to be passed for rolling window period in calculations.

This ensures that pre-processing is reduced for the data thus improving the performance.

The scheme data and the benchmark data can either be a pandas dataframe or list of tuples: (date, nav).

Also you can use it to calculate xirr:

>>> import finance_calculator as fc
>>> cashflow_data = [
    (datetime.date(2020, 3, 1), 10000),
    (datetime.date(2020, 4, 1), 10000),
    (datetime.date(2020, 5, 1), 10000),
    (datetime.date(2020, 6, 1), 10000),
    (datetime.date(2020, 7, 1), 10000),
    (datetime.date(2020, 8, 1), -60000),
]
>>> xirr = fc.get_xirr(cashflow_data)