optimus.backtesting package

Submodules

optimus.backtesting.data module

class optimus.backtesting.data.Universe(selection: optimus.backtesting.selection.UniverseSelection, indicator: optimus.backtesting.indicator.Indicator, auto_run: bool = True)

Bases: optimus.backtesting.api_key.ApiKey

Generate an Universe for backtesting

Parameters

  • selection (UniverseSelection) – Stocks, Warrants, or Indices

  • indicator (Indicator) – Indicator

  • auto_run (bool) – Computation, defaults to True

property df

Translate into DataFrame

Returns

pd.DataFrame

df_multi_period(*args)

Slicing DataFrame into multi period time frame.

Usage:

1
2
3
4
first_date = datetime(2015, 1, 1)
second_date = datetime(2016, 1, 1)
third_date = datetime(2017, 1, 1)
df_multi_period(first_date, second_date, third_date)
Parameters

args (python.datetime.datetime) – datetime(2015, 1, 1), datetime(2016, 1, 1), datetime(2017, 1, 1)

Returns

Lists of DataFrame according to specified time frame

df_period_shifting(date: datetime.datetime, shift: int, multiplier: int = 1)

Slicing DataFrame into multi period time frame with shifting

Usage:

1
2
date = datetime(2016, 1, 4)
df_period_shifting(date, 10, 2)

Results:

DataFrame 1

date

…

2016-01-04

…

2016-01-05

…

… 7 more rows …

…

2016-01-15

…

2016-01-18

…
DataFrame 2

date

…

2016-01-18

…

2016-01-19

…

… 7 more rows …

…

2016-02-02

…

2016-02-03

…
Parameters

  • date (python.datetime.datetime) – datetime(2016, 1, 4)

  • shift (int) – period shifting, can be positive (look forward), or negative (look backward)

  • multiplier – shift multiplier, repeat shifting, defaults to 1

Returns

Lists of DataFrame shifted

plot_heatmap(path, metric, rows, cols, x_label, y_label, width=15, height=10, save_figure=False, name='heatmap.png')

Plot heatmap according to csv file from two changing parameters

Parameters

  • path (path) – path of objective function / performance metric

  • metric (string) – name of metric in csv file

  • rows (list) – list containing values of row parameter

  • cols (list) – list containing values of col parameter

  • x_label (string) – name of x-axis label

  • y_label (string) – name of y-axis label

  • width (int) – width of figure

  • height (int) – height of figure

  • save_figure (bool) – save figure as png

  • name (string) – name of figure

optimus.backtesting.performance module

class optimus.backtesting.performance.Performance(portfolio: optimus.backtesting.portfolio.Portfolio, show: bool = True)

Bases: object

Performance module for backtesting

Parameters

  • portfolio (Portfolio) – Portfolio

  • show (bool) – Show graph or not

property average_losing_trade

Average loss of all losing trades. Losing trades is defined as < 0

property average_number_of_bars_in_losers

Average number of days for all round losing trades. Losing trades is defined as < 0

property average_number_of_bars_in_winners

Average number of days for all round winning trades. Winning trades is defined as > 0

property average_trade

Average profit of all trades

property average_winning_trade

Average profit of all winning trades. Winning trades is defined as > 0

generate_all_files(strategy_name, code_path, version, plot_image=True)

Generate performances for all files

Parameters

  • strategy_name (string) – folder name of strategy

  • code_path (file path) – path of code file

  • version (string) – version of the strategy

  • plot_image (bool) – save images into folder

property gross_loss

Sum of all loss. Loss is defined as value < 0

property gross_profit

Sum of all profit. Profit is defined as value > 0

property largest_losing_trade

Largest loss among all losing trades. Losing trades is defined as < 0

property largest_winning_trade

Largest profit among all winning trades. Winning trades is defined as > 0

property max_daily_drawdown

Daily drawdown means difference between Cumulative net profit and Maximum Cumulative net profit

Maximum daily drawdown means Largest drawdown or decline in the equity curve of a trading strategy from an equity peak

Measures the downside of the system but ignores the upside potential

Maximum daily drawdown is in % to cater for the size of portfolio throughout time

Look for smaller maximum daily drawdown

property maximum_intraday_drawdown

Largest possible net loss in the system throughout the trading period.

property maximum_intraday_profit

Largest possible net profit in the system throughout the trading period.

property net_profit

Sum of all profit and loss at the end date

Measures the upside of the system but ignores the distribution of trades and downside risk

Look for larger net profit

property number_losing_trades

Number of complete round losing trades. Losing trades is defined as < 0

property number_winning_trades

Number of complete round winning trades. Winning trades is defined as > 0

objective_function(need_print=True)
objective_to_csv(strategy_name, file_path)

Convert transactions to csv file

Parameters

  • strategy_name (str) – Name of the strategy

  • file_path (path) – File path

property percent_profitable
\[\dfrac{Number\ winning\ trades}{Total\ number\ of\ trades}\]

Look for percent profitable above 50%, especially in mean reversion strategy

performance_metric(need_print=True)
performance_to_csv(strategy_name, file_path)

Convert transactions to csv file

Parameters

  • strategy_name (str) – Name of the strategy

  • file_path (path) – File path

plot_objective_function(save_figure=False)
plot_performance_metric(save_figure=False)
property profit_factor
\[\dfrac{Gross\ profit}{Gross\ loss}\]

Look for factor > 1, for all type of system

property prom

PROM assumes trading strategy will win less and lose more in real-time trading.

\[PROM = \dfrac{(AW×(WT-\sqrt{WT}))-(AL×(LT-\sqrt{LT}))}{Initial\ Capital}\]

WT = Number of Wins

AW = Average Wins

LT = Number of Losses

AL = Average Loss

Look for higher PROM.

property prom_less

PROM minus biggest win removes the largest single profitable trade from the gross profit.

Look for higher PROM minus biggest win

property ratio_average_win_over_average_loss

Average winning trade / average losing trade in absolute term

Look for ratio above 1, especially in momentum trend following system

property sharpe_ratio

Formula:

\[\dfrac{Mean[Asset\ return - Risk\ free\ return]}{Standard\ deviation\ of\ asset\ excess\ return}\]

We take 2.8% for Risk free return, 10Y MGS rate.

Look for Sharpe ratio above 1

property sqn

SQN is trademark of Van Tharp, based on R-multiples.

\[R-multiples = \dfrac{Exit\ price - Entry\ price}{Entry\ price - Cut\ loss\ price}\]
\[SQN = \dfrac{Mean\ of\ R-multiples}{Standard\ Deviation\ of\ R-multiples}×{\sqrt{Number\ of\ Trades}}\]

  • Number of Trades we take min (number of trades, 100 ) because too many trades will overestimate the quality of the system

Look for sqn above 2. Above 7 is considered holy grail.

property total_number_of_trades

Number of complete round trades. Round trades defined as opening position with a Buy and closing postion with a Sell

transactions_to_csv(path)

Convert transactions to csv file

Parameters

path – File path

optimus.backtesting.portfolio module

class optimus.backtesting.portfolio.Portfolio(df: pandas.core.frame.DataFrame, equity_model: optimus.backtesting.selection.EquityModel = <EquityModel.CORE_EQUITY: 1>, position_sizing: optimus.backtesting.selection.PositionSizing = <PositionSizing.EQUAL_UNIT: 1>, initial_capital: int = 100000, size: int = 10, take_profit: Optional[float] = None, cut_loss: float = 0.08, commission: bool = False, brokerage_fee: float = 0.08)

Bases: object

Constructs a Portfolio

Parameters

  • df (pd.DataFrame) – Dataframe used to calculate portfolio value

  • equity_model (EquityModel) – Must be an instance of EquityModel

  • position_sizing (PositionSizing) – Must be an instance of PositionSizing

  • initial_capital (int) – Initial capital allocated for the portfolio

  • size (int) – Portfolio size

  • take_profit (float) – Take profit percentage

  • cut_loss (float) – Cut loss percentage

  • commission (bool) – Calculate commission into transactions

  • brokerage_fee (float) – Brokerage fee calculated in percentage of total value traded

allocation() → float

Capital allocated for transactions

Returns

Capital allocated

property brokerage_fee

Brokerage fee calculated in percentage of total value traded.

Returns

Float

buy(ticker, date, price, index, tp_price=None, cl_price=None, risk_percent=0.02)

Create a buy transaction. If successful ticker will be added into portfolio and added into transaction list.

Parameters

  • ticker – ticker symbol

  • date – order date

  • price – order price

  • index – index of open date, used to compute holding days

  • tp_price – Custom take profit price

  • cl_price – Custom cut loss price

  • risk_percent – percentage of equity to risk (percent volatility & percent risk model)

calculate_commission(unit, price)

Calculate commission given unit size and price

calculate_position_size(date, ticker, price, risk_percent, cl_price)

Calculate position size based on equity model and position size model

property capital

Current capital

Returns

Integer/ Float

clear()

Clear portfolio

property commission

Calculate commission into transactions.

Returns

Bool

property cut_loss

Cut loss percentage set by user.

Returns

Float

free() → int

Slots available in portfolio.

Returns

Free slot in portfolio

get_equity(date)

Get equity based on different models

holding() → List[Dict]

Portfolio holding

Returns

List of portfolio holdings

property initial_capital

Initial capital set by user

Returns

Integer

list() → List[str]

Portfolio holding ticker list

Returns

List of portfolio holdings’ ticker list

portfolio_value(date)

Calculate portfolio value given specific date

sell(ticker, reason, date, price, index)

Create a sell transaction. If successful ticker will be removed from portfolio and added into transaction list.

Parameters

  • ticker – ticker symbol

  • reason – reason for selling

  • date – order date

  • price – order price

  • index – index of close date, used to compute holding days

set_cut_loss_price(ticker, cl_price)

Update the cut loss price of a ticker in portfolio holding

Parameters

  • ticker – ticker symbol

  • cl_price – cut loss price

property size

Size of the portfolio.

Returns

Integer

property take_profit

Take profit percentage set by user.

Returns

Float

property transactions

Transactions list generated.

Returns

List

optimus.backtesting.indicator module

class optimus.backtesting.indicator.Indicator

Bases: object

Indicator class.

adxr(params: Optional[List[List[int]]] = None)

Average Directional Index Rating

Series named as ADX_[ADX Smoothing]_[DI Length], ADXR_[ADX Smoothing]_[DI Length]_[Lookback Period]

Parameters

params (List[List[int, int, int]]) – List[List[ADX Smoothing, DI Length, Lookback Period]], defaults to [[14, 14, 14]]

all()

Includes all indicators

SMA, EMA, VMA, WMA, HMA, On Balance Volume, Bollinger Band, Stochastic Oscillator, MACD, Pivot, Momentum, RSI, Keltner Channel, Biggest Drop

atr(params: Optional[List[int]] = None)

Average True Range

Series named as ATR_[Period]. E.g. ATR_14.

Parameters

params (List[int]) – Arrays of ATR period, defaults to [14]

average_daily_turnover(params: Optional[List[int]] = None)

Average daily turnover

Series named as average_daily_turnover_[Period]. E.g. average_daily_turnover_20.

Parameters

params (List[int]) – Arrays of average daily turnover period, defaults to [20]

biggest_drop(params: Optional[List[int]] = None)

Biggest Drop

Series named as biggest_drop_[Period]. E.g. biggest_drop_3.

Parameters

params (List[int]) – Arrays of biggest_drop period, defaults to [3]

bollinger_band(params: Optional[List[List[int]]] = None)

Bollinger Band

Series named as LowerBand_[MA]_[STD], MiddleBand_[MA]_[STD], UpperBand_[MA]_[STD], BbWidth_[MA]_[STD]. E.g. LowerBand_20_2, MiddleBand_20_2, UpperBand_20_2, BbWidth_20_2

Parameters

params (List[List[int, int]]) – List[List[MA, STD]], defaults to [[20, 2], [20, 3]]

candle_bearish_abandoned_baby(params: Optional[List[int]] = None)

Candlestick pattern - bearish abandoned baby

Series named as candle_bearish_abandoned_baby_[Period]. E.g. candle_bearish_abandoned_baby_3

Parameters

params (List[int]) – Arrays of candle_bearish_abandoned_baby period, defaults to [3]

candle_bearish_engulfing(params: Optional[List[int]] = None)

Candlestick pattern - bearish engulfing

Series named as candle_bearish_engulfing_[Period]. E.g. candle_bearish_engulfing_3

Parameters

params (List[int]) – Arrays of candle_bearish_engulfing period, defaults to [3]

candle_bearish_harami(params: Optional[List[int]] = None)

Candlestick pattern - bearish harami

Series named as candle_bearish_harami_[Period]. E.g. candle_bearish_harami_3

Parameters

params (List[int]) – Arrays of candle_bearish_harami period, defaults to [3]

candle_bullish_abandoned_baby(params: Optional[List[int]] = None)

Candlestick pattern - bullish abandoned baby

Series named as candle_bullish_abandoned_baby_[Period]. E.g. candle_bullish_abandoned_baby_3

Parameters

params (List[int]) – Arrays of candle_bullish_abandoned_baby period, defaults to [3]

candle_bullish_engulfing(params: Optional[List[int]] = None)

Candlestick pattern - bullish engulfing

Series named as candle_bullish_engulfing_[Period]. E.g. candle_bullish_engulfing_3

Parameters

params (List[int]) – Arrays of candle_bullish_engulfing period, defaults to [3]

candle_bullish_harami(params: Optional[List[int]] = None)

Candlestick pattern - bullish harami

Series named as candle_bullish_harami_[Period]. E.g. candle_bullish_harami_3

Parameters

params (List[int]) – Arrays of candle_bullish_harami period, defaults to [3]

candle_evening_star(params: Optional[List[int]] = None)

Candlestick pattern - evening star

Series named as candle_evening_star_[Period]. E.g. candle_evening_star_3

Parameters

params (List[int]) – Arrays of candle_evening_star period, defaults to [3]

candle_hammer(params: Optional[List[int]] = None)

Candlestick pattern - hammer

Series named as candle_hammer_[Period]. E.g. candle_hammer_3

Parameters

params (List[int]) – Arrays of candle_hammer period, defaults to [3]

candle_hanging_man(params: Optional[List[int]] = None)

Candlestick pattern - hanging man

Series named as candle_hanging_man_[Period]. E.g. candle_hanging_man_3

Parameters

params (List[int]) – Arrays of candle_hanging_man period, defaults to [3]

candle_morning_star(params: Optional[List[int]] = None)

Candlestick pattern - morning star

Series named as candle_morning_star_[Period]. E.g. candle_morning_star_3

Parameters

params (List[int]) – Arrays of candle_morning_star period, defaults to [3]

donchian_channels(params: Optional[List[int]] = None)

Donchian channels

Series named as DC_Lower_[period], DC_Upper_[period], DC_Basis_[period], DC_Lower_last_[period], DC_Upper_last_[period]. E.g. DC_Upper_20.

DC_Lower_last_[period]: Date of candle that created donchian low DC_Upper_last_[period]: Date of candle that created donchian high

Parameters

params (List[int]) – Arrays of donchian channels period, defaults to [20]

ema(params: Optional[List[int]] = None)

Exponential Moving Average

Series named as EMA_[param]. E.g. EMA_4, EMA_5.

Parameters

params (List[int]) – Arrays of EMA period, defaults to [4, 5, 10, 13, 20, 26, 50, 150]

fibonacci_pullback(params: Optional[List[int]] = None)

Fibonacci Pullback

Series named as pullback_[Period]. E.g. pullback_14.

Parameters

params (List[int]) – Arrays of Donchian channels period to calculate fibonacci pullback, defaults to [14]

fractal(params: Optional[List[int]] = None)

William fractal indicator

Series named as bullish_fractal_count_[period], bearish_fractal_count_[period].

Parameters

params (List[int]) – Arrays of fractal period, defaults to [2,4]

fractal_breakout(params: Optional[List[int]] = None)

Fractal breakout indicator Calculate breakout percentage from fractal high

Series named as breakout%_[period].

Parameters

params (List[int]) – Arrays of fractal breakout period, defaults to [3]

gainer_loser(enabled: bool = True)

Gainer loser indicator

Series named as gainer_loser

Parameters

enabled (bool) – defaults to True

high_low(params: Optional[List[int]] = None)

High low indicator

Series named as high_[period], low_[period]. E.g. high_21, low_21. Difference with Donchian channel: Before specified period - Calculate cumulative high / low After specified period - Calculate rolling high / low

Parameters

params (List[int]) – List[high_low_periods], defaults to [21, 63, 126, 252]

hma(params: Optional[List[int]] = None)

Hull Moving Average

Series named as HMA_[param]. E.g. HMA_5, HMA_10.

Parameters

params (List[int]) – Arrays of HMA period, defaults to [5, 10]

keltner(params: Optional[List[List[Union[int, float]]]] = None)

Keltner Channel

Series named as KelChU_[MA]_[ATR]_[Multiplier], KelChD_[MA]_[ATR]_[Multiplier], squeeze_[MA]_[ATR]_[Multiplier]. E.g. KelChU_[20]_[20]_[1.5], KelChD_[20]_[20]_[1.5], squeeze_[20]_[20]_[1.5]

Parameters

params (List[List[Union[int, float], Union[int, float], Union[int, float]]]) – List[List[MA, ATR, Multiplier]], defaults to [[20, 20, 1.5]]

length()

Indicators size

macd(params: Optional[List[List[int]]] = None)

Moving Average Convergence Divergence

Series named as MACD_[fast]_[slow]_[ema], MACDsign_[fast]_[slow]_[ema], MACDdiff_[fast]_[slow]_[ema]. E.g. MACD_12_26_9.

Parameters

params (List[List[int, int, int]]) – List[List[SMA Fast, SMA Slow, EMA]], defaults to [[12, 26, 9]]

momentum(params: Optional[List[int]] = None, weightage: Optional[List[float]] = None)

Proprietary momentum indicator

Series named as momentum

Parameters

  • params (List[float]) – 4 windows, defaults to [63, 126, 189, 252]

  • weightage – Weightage of each window, summation of it must be equals to 1, defaults to [0.4, 0.2, 0.2, 0.2]

on_balance_volume(enabled: bool = True)

On balance volume indicator

Series named as OBV

Parameters

enabled (bool) – defaults to True

parabolic_sar(params: Optional[List[List[Union[int, float]]]] = None)

Parabolic SAR

Series named as SAR_[start]_[increment]_[maxValue], SAR_status_[start]_[increment]_[maxValue]. E.g. SAR_0.02_0.02_0.2.

Parameters

params (List[List[Union[int, float], Union[int, float], Union[int, float]]]) – List[List[Start, Increment, Max value]], defaults to [[0.02, 0.02, 0.2]]

percentage_change(enabled: bool = True)

Percentage change indicator (today against yesterday)

Series named as percentage_change

Parameters

enabled (bool) – defaults to True

pivot(enabled: bool = True)

Proprietary pivot indicator

Series named as upPivot, downPivot

Parameters

enabled (bool) – defaults to True

roc(params: Optional[List[int]] = None)

Rate of change

Series named as ROC_[Period]. E.g. ROC_9.

Parameters

params (List[int]) – Arrays of roc period, defaults to [9]

rsi(params: Optional[List[int]] = None)

Relative Strength Index

Series named as RSI_[period]. E.g. RSI_14

Parameters

params (List[int]) – Arrays of RSI period, defaults to [14]

sma(params: Optional[List[int]] = None)

Simple Moving Average.

Series named as SMA_[param]. E.g. SMA_5, SMA_10.

Parameters

params (List[int]) – Arrays of SMA period, defaults to [5, 10, 13, 20, 26, 50, 150, 200]

sma_ema(params: Optional[List[List[Union[str, int]]]] = None)

Sma_ema indicator

Series named as breadth_sma5_[confidence_level], breadth_ema4_[confidence_level]. E.g. breadth_sma5_75

Confidence level is restricted to 75, 80, 85, 90.

Parameters

params (List[List[Union[str, int]]]) – List[List[Union[sma5/ema4, confidence_level]]]

smart_money(params: Optional[List[List[Union[int, float]]]] = None)

Optimus smart money indicator

Series named as: retailer_[banker_base]_[banker_rsi_period]_[hot_money_rsi_base]_[hot_money_rsi_period]_[sensitivity_banker]_[sensitivity_hot_money]_[banker_ma]_[retailer_ma] hotMoney_[banker_base]_[banker_rsi_period]_[hot_money_rsi_base]_[hot_money_rsi_period]_[sensitivity_banker]_[sensitivity_hot_money]_[banker_ma]_[retailer_ma] banker_[banker_base]_[banker_rsi_period]_[hot_money_rsi_base]_[hot_money_rsi_period]_[sensitivity_banker]_[sensitivity_hot_money]_[banker_ma]_[retailer_ma] bankerMA_[banker_base]_[banker_rsi_period]_[hot_money_rsi_base]_[hot_money_rsi_period]_[sensitivity_banker]_[sensitivity_hot_money]_[banker_ma]_[retailer_ma] retailerMA_[banker_base]_[banker_rsi_period]_[hot_money_rsi_base]_[hot_money_rsi_period]_[sensitivity_banker]_[sensitivity_hot_money]_[banker_ma]_[retailer_ma] E.g. retailer_50_50_30_40_1.5_0.7_10_10.

Parameters

params (List[List[int, int, int, int, Union[int, float], Union[int, float], int, int]]) – List[List[banker_base, banker_rsi_period, hot_money_rsi_base, hot_money_rsi_period, sensitivity_banker, sensitivity_hot_money, banker_ma, retailer_ma]], defaults to [[50_50_30_40_1.5_0.7_10_10]]

smma(params: Optional[List[int]] = None)

Smoothed Moving Average

Series named as SMMA_[param]. E.g. SMMA_4, SMMA_5.

Parameters

params (List[int]) – Arrays of SMMA period, defaults to [50, 200]

stc(params: Optional[List[List[int]]] = None)

Schaff Trend Cycle

Series named as STC_[fast]_[slow]_[percK]_[percD]_[stc].

Parameters

params (List[List[int, int, int, int, int]]) – List[List[EMA Fast, EMA Slow, Stoch %K Length, Stoch %D Length, STC Length]], defaults to [[23, 50, 10, 3, 3]]

stochastic(params: Optional[List[List[int]]] = None)

Stochastic Oscillator

Series named as SOk_[Period]_[MA], SOd_[Period]_[MA]. E.g. SOk_14_3, SOd_14_3.

Parameters

params (List[List[int, int]]) – List[List[Period, MA]], defaults to [[14, 3]]

trend(params: Optional[List[List[Union[int, float]]]] = None)

Trend

Series named as Trend_[MA length]_[Stdev multiple]_[Strong Stdev multiple]_[Lookback Period]

Parameters

params (List[List[Union[int, float], Union[int, float], Union[int, float], Union[int, float]]]) – List[List[MA length, Stdev multiple, Strong Stdev multiple, Lookback Period]], defaults to [[150, 0.06, 0.16, 15]]

vma(params: Optional[List[int]] = None)

Volume Moving Average

Series named as VMA_[param]. E.g. VMA_5, VMA_10.

Parameters

params (List[int]) – Arrays of VMA period, defaults to [5, 10, 20]

volume_average(params: Optional[List[int]] = None)

Volume average

Series named as volume_average_[param]. E.g. volume_average_10.

Parameters

params (List[int]) – Arrays of volume average period, defaults to [10]

wma(params: Optional[List[int]] = None)

Weighted Moving Average

Series named as WMA_[param]. E.g. WMA_5, WMA_10.

Parameters

params (List[int]) – Arrays of WMA period, defaults to [5, 10]

ytd_change(enabled: bool = True)

Year to date change (in percent) indicator

Series named as ytd_change

Parameters

enabled (bool) – defaults to True

optimus.backtesting.selection module

class optimus.backtesting.selection.Action(value)

Bases: enum.Enum

An enumeration.

BUY = 1
SELL = 2
class optimus.backtesting.selection.EquityModel(value)

Bases: enum.Enum

An enumeration.

CORE_EQUITY = 1
REDUCED_TOTAL_EQUITY = 3
TOTAL_EQUITY = 2
class optimus.backtesting.selection.PositionSizing(value)

Bases: enum.Enum

An enumeration.

EQUAL_UNIT = 1
PERCENT_RISK = 3
PERCENT_VOLATILITY = 2
class optimus.backtesting.selection.Reason(value)

Bases: enum.Enum

An enumeration.

CL = 3
ENTRY = 1
EXIT = 2
TP = 4
class optimus.backtesting.selection.UniverseSelection(value)

Bases: enum.Enum

An enumeration.

INDICES = 3
STOCKS = 1
WARRANTS = 2

optimus.backtesting.walk_forward module

class optimus.backtesting.walk_forward.WalkForward(dataframe, data_set, split_ratio_train, split_ratio_test, objective_function, portfolio, backtest, *args)

Bases: object

Walk forward optimization for multiple dataframe.

Parameters

  • dataframe (pandas.DataFrame) – Dataframe of stock data

  • data_set (int) – Number of data set to be split

  • split_ratio_train (int) – Split ratio of train set

  • split_ratio_test (int) – Split ratio of test set

  • objective_function (str) – Objective function to optimize

  • portfolio (Portfolio) – Portfolio of test set

  • backtest (types.FunctionType) – Backtest function

  • *args (args) – values for each parameter to be optimized

backtest_test_set(complete_search=True, strategy_name=None, file_path=None)

Start backtest for test set using best params of train set (complete search or prioritized search), store results in folder

Parameters

  • complete_search (boolean) – true if train set is backtested using complete search, false if train set is backtested using prioritized search

  • strategy_name (string) – folder name of strategy

  • file_path (file path) – file path for csv file of complete search train set

backtest_train_set_complete()

Start backtest for train set using complete search.

backtest_train_set_prioritized()

Start backtest for train set using prioritized search.

optimus.backtesting.tick_data module

class optimus.backtesting.tick_data.TickData

Bases: object

Generate a DataFrame for tick data

get_TA(timeframe, indicator: optimus.backtesting.indicator.Indicator)

Get technical indicators for a timeframe.

Parameters

  • timeframe (string) – Desired timeframe

  • indicator (Indicator) – Desired indicators

get_timeframe(timeframe, date)

Get a timeframe given a date.

Parameters

  • timeframe (string) – Desired timeframe

  • date (string) – Desired date

resample(timeframe)

Resample intraday timeframe from tick data dataframe and save into file

Intraday timeframe examples: min, H

Parameters

timeframe (string) – Desired timeframe

Module contents