High Frequency Portfolio Analytics by PortfolioEffect

R interface to PortfolioEffect cloud service for backtesting high frequency trading (HFT) strategies, intraday portfolio analysis and optimization. Includes auto-calibrating model pipeline for market microstructure noise, risk factors, price jumps/outliers, tail risk (high-order moments) and price fractality (long memory). Constructed portfolios could use client-side market data or access HF intraday price history for all major US Equities. See < https://www.portfolioeffect.com/> for more information on the PortfolioEffect high frequency portfolio analytics platform.


PortfolioEffectHFT v1.8 (Release date: 2017-03-21)


  • Fixed a problem in plot functions.
  • Fixed a problem in 'util_getComputeTime' functions.
  • Added "font_size","line_size","bw","axis.text.size" and "title.size" output option to "plot" function "portfolio" and "position" class.
  • Included arguments was changed in the "portfolio_availableSymbols" function.

PortfolioEffectHFT v1.7 (Release date: 2016-07-15)


  • Calculation logic was changed to always return metric classes to allow lazy evaluation and auto-batching calculations.
  • The following metric functions was renamed: portfolio_addPosition -> position_add portfolio_alpha -> alpha_exante portfolio_beta -> beta portfolio_calmarRatio -> calmar_ratio portfolio_create -> portfolio_create portfolio_cumulant -> cumulant portfolio_CVaR -> expected_shortfall portfolio_downCaptureRatio -> down_capture_ratio portfolio_downNumberRatio -> down_number_ratio portfolio_downPercentageRatio -> down_percentage_ratio portfolio_downsideVariance -> downside_variance portfolio_expectedDownsideReturn -> expected_downside_return portfolio_expectedReturn -> expected_return portfolio_expectedUpsideReturn -> expected_upside_return portfolio_fractalDimension -> fractal_dimension portfolio_gainLossVarianceRatio -> gain_loss_variance_ratio portfolio_gainVariance -> gain_variance portfolio_hurstExponent -> hurst_exponent portfolio_informationRatio -> information_ratio portfolio_jensensAlpha -> alpha_jensens portfolio_kurtosis -> kurtosis portfolio_lossVariance -> loss_variance portfolio_maxDrawdown -> max_drawdown portfolio_modifiedSharpeRatio -> mod_sharpe_ratio portfolio_moment -> moment portfolio_omegaRatio -> omega_ratio portfolio_pdf -> dist_density portfolio_profit -> profit portfolio_rachevRatio -> rachev_ratio portfolio_removePosition -> position_remove portfolio_return -> log_return portfolio_sharpeRatio -> sharpe_ratio portfolio_skewness -> skewness portfolio_sortinoRatio -> sortino_ratio portfolio_starrRatio -> starr_ratio portfolio_startBatch -> portfolio_startBatch portfolio_symbols -> position_list portfolio_treynorRatio -> treynor_ratio portfolio_txnCosts -> txn_costs portfolio_upCaptureRatio -> up_capture_ratio portfolio_upNumberRatio -> up_number_ratio portfolio_upPercentageRatio -> up_percentage_ratio portfolio_upsideDownsideVarianceRatio -> upside_downside_variance_ratio portfolio_upsideVariance -> upside_variance portfolio_value -> value portfolio_VaR -> value_at_risk portfolio_variance -> variance position_alpha -> alpha_exante position_beta -> beta position_calmarRatio -> calmar_ratio position_correlation -> correlation position_correlationMatrix -> correlation position_covariance -> covariance position_covarianceMatrix -> covariance position_cumulant -> cumulant position_CVaR -> expected_shortfall position_downCaptureRatio -> down_capture_ratio position_downNumberRatio -> down_number_ratio position_downPercentageRatio -> down_percentage_ratio position_downsideVariance -> downside_variance position_expectedDownsideReturn -> expected_downside_return position_expectedReturn -> expected_return position_expectedUpsideReturn -> expected_upside_return position_fractalDimension -> fractal_dimension position_gainLossVarianceRatio -> gain_loss_variance_ratio position_gainVariance -> gain_variance position_hurstExponent -> hurst_exponent position_informationRatio -> information_ratio position_jensensAlpha -> alpha_jensens position_kurtosis -> kurtosis position_lossVariance -> loss_variance position_maxDrawdown -> max_drawdown position_modifiedSharpeRatio -> mod_sharpe_ratio position_moment -> moment position_omegaRatio -> omega_ratio position_pdf -> dist_density position_price -> price position_profit -> profit position_quantity -> quantity position_rachevRatio -> rachev_ratio position_return -> log_return position_returnAutocovariance -> return_autocovariance position_returnJumpSize -> return_jump_size position_setQuantity -> set_quantity position_sharpeRatio -> sharpe_ratio position_skewness -> skewness position_sortinoRatio -> sortino_ratio position_starrRatio -> starr_ratio position_treynorRatio -> treynor_ratio position_txnCosts -> txn_costs position_upCaptureRatio -> up_capture_ratio position_upNumberRatio -> up_number_ratio position_upPercentageRatio -> up_percentage_ratio position_upsideDownsideVarianceRatio -> upside_downside_variance_ratio position_upsideVariance -> upside_variance position_value -> value position_VaR -> value_at_risk position_variance -> variance position_weight -> weight
  • Added "metric" class with class methods "compute", "plot", "+", "-", "*", ":". This class can be obtained using any metric function ("beta", "alpha", ...) or "create_metric" function.
  • Added "position" class with class methods "show", "plot". This class can be obtained using "position_add" or "portfolio_getPosition" functions.
  • Added "forecast" class with class methods "forecast_input", "forecast_apply". This class can be obtained using "forecast_builder" function. Forecasting supports linear heterogeneous autoregression (HAR) in a rolling window.
  • The following functions was replaced by "optimization_constraint" with other included function arguments: optimization_constraint_allWeights optimization_constraint_beta optimization_constraint_CVaR optimization_constraint_expectedReturn optimization_constraint_modifiedSharpeRatio optimization_constraint_portfolioValue optimization_constraint_return optimization_constraint_sharpeRatio optimization_constraint_starrRatio optimization_constraint_sumOfAbsWeights optimization_constraint_VaR optimization_constraint_variance optimization_constraint_weight
  • Included arguments was changed in the "optimization_goal" function.
  • Auto-batching of lazy metrics is enabled by default. The following batch functions have been removed: portfolio_startBatch portfolio_endBatch

PortfolioEffectHFT v1.6 (Release date: 2016-07-15)


  • Fixed a problem in vignettes that was preventing the package from building on Linux

PortfolioEffectHFT v1.5 (Release date: 2016-02-01)


  • Switched to using ggplot 2.0
  • Added "bw" (black and white) output option to util_summary()
  • Added "resultsNAFilter" to portfolio_settings()

PortfolioEffectHFT v1.4 (Release date: 2015-12-17)


  • Improvements in estimates precision
  • Fixed a number of bugs that occurred only under high load
  • Improvements in server-side data caching
  • Output "NaN" for missing values, computational errors, warm-up periods and possible artifacts

PortfolioEffectHFT v1.3 (Release date: 2015-11-08)


  • Added portfolio_availableSymbols() to list all available server-side instruments
  • Added "fractalPriceModel" to portfolio_settings() method to turn on/off fractal price model.
  • Confidence interval parameter changed meaning from tail probability to 1 - tail probability (e.g.from 0.05 to 0.95)
  • Default value of "driftTerm" in portfolio_settings() is now "FALSE"
  • Fixed errors in portfolio_startBatch() and portfolio_endBatch() - both methods are now working

PortfolioEffectHFT v1.2 (Release date: 2015-09-29)

First commit.

New functionality:

  • Auto-calibrating model pipeline for market microstructure noise, risk factors, price jumps/outliers, tail risk (high-order moments), price fractality (long memory) that was built to give tick-resolution analytics.
  • Over 40+ portfolio and position-level metrics of intraday risk and performance from modern and post-modern portfolio theory.
  • Single-period constraint portfolio optimization (classic Markowitz and extensions for tail risk) with scalar, vector-based and user-defined functional constraints
  • Multi-period constraint portfolio optimization that accounts for previous portfolio rebalancing (trading strategy optimization).
  • Transactional costs were also implemented in this release.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.8 by Andrey Kostin, 3 years ago


Browse source code at https://github.com/cran/PortfolioEffectHFT

Authors: Andrey Kostin [aut, cre] , Aleksey Zemnitskiy [aut] , Oleg Nechaev [aut] , Craig Otis and others [ctb, cph] (OpenFAST library) , Daniel Lemire , Muraoka Taro and others [ctb, cph] (JavaFastPFOR library) , Joe Walnes , Jorg Schaible and others [ctb, cph] (XStream library) , Dain Sundstrom [ctb, cph] (Snappy library) , Extreme! Lab , Indiana University [ctb, cph] (XPP3 library) , The Apache Software Foundation [ctb, cph] (Apache Log4j and Commons Lang libraries) , Google , Inc. [ctb, cph] (GSON library) , Free Software Foundation [ctb, cph] (GNU Trove and GNU Crypto libraries)

Documentation:   PDF Manual  

Task views: Empirical Finance

GPL-3 license

Imports methods, rJava, grid, zoo

Depends on ggplot2

Suggests testthat

System requirements: Java (>= 1.7)

Depended on by PortfolioEffectEstim.

See at CRAN