TimeseriesAnalysis#

class qf_lib.analysis.timeseries_analysis.timeseries_analysis.TimeseriesAnalysis(returns_timeseries: QFSeries, frequency: Frequency)[source]#

Bases: TimeseriesAnalysisDTO

Used for analysing a timeseries of returns. Calculates and aggregates different statistics of the timeseries, It contains the following fields:

  • returns_tms - series of simple returns

  • frequency

  • start_date

  • end_date

  • total_return

  • cagr - annualised return

  • annualised_vol

  • annualised_upside_vol

  • annualised_downside_vol

  • sharpe_ratio

  • omega_ratio

  • calmar_ratio

  • gain_to_pain_ratio

  • sortino_ratio

  • cvar - 5% CVaR expressed related to the specified frequency

  • annualised_cvar - annualised 5% CVaR

  • max_drawdown - maximum drawdown

  • avg_drawdown - average of the whole underwater chart

  • avg_drawdown_duration - average duration of a drawdown

  • best_return

  • worst_return

  • vavg_positive_return

  • avg_negative_return

  • skewness

  • kurtosis

  • kelly

Parameters:

  • returns_timeseries (QFSeries) – Analysed timeseries. It should be PriceSeries, SimpleReturnSeries or LogReturnSeries

  • frequency (Frequency) – Corresponds to the frequency od data samples in the seres.

get_measures() → List[str][source]#

Returns a list of all measures (values) represented as strings

get_nice_names() → List[str][source]#

Returns a list of long, nice names of all the measures

get_short_names() → List[str][source]#

Returns a list of short names of all the measures

get_units() → List[str][source]#

Returns a list of all units of all measures (values)

populate_table(table: Table, name=None) → None[source]#

Adds the data calculated in this analysis to the specified table. The table may be brand new or contain other analyses of the same kind.

Parameters:

  • table – The table to add the data to.

  • name – Name to give this analysis in the columns.

static table_for_df(df: QFDataFrame, frequency: Frequency = Frequency.DAILY) → str[source]#

Returns a table similar to the one below:

Analysed period: start_date - end_date, using frequency data Name total_ret cagr vol up_vol down_vol … Asset1 63.93 28.27 19.15 14.06 14.35 … Asset2 66.26 29.19 20.74 14.86 15.54 … Asset3 66.26 29.19 20.74 14.86 15.54 … … … … … … … …

Parameters:

  • df – DataFrame of returns or prices of assets to be analysed

  • frequency – (optional) frequency of the returns or price sampling in the DataFrame. By default daily frequency is used

static values_in_table(ta_collection: TimeseriesAnalysis | Sequence[TimeseriesAnalysis], asset_names: None | str | Sequence[str] = None) → str[source]#

Returns a string with all the measures in a form of table of the following format:

Asset1 Asset2 …

Nice_name1 value11 value21 … unit1 Nice_name2 value12 value22 … unit2 … … … … unitI Nice_nameN value12 value22 … unitN

Parameters:

  • ta_collection – single TimeseriesAnalysis object or a collection of TimeseriesAnalysis objects

  • asset_names – names of assets corresponding to objects in ta_collection