Skip to content

Latest commit

 

History

History
266 lines (182 loc) · 10.9 KB

CHANGELOG.rst

File metadata and controls

266 lines (182 loc) · 10.9 KB

Changelog

https://img.shields.io/static/v1?label=Sponsor&message=%E2%9D%A4&logo=GitHub&color=%23fe8e86
Buy Me a Coffee at ko-fi.com

Version 6.3.0

  • Add new functions to calculate the number of effective assets (NEA) and the average centrality of the portfolio.
  • Add the possibility to use neighborhood and cluster network constraints at the same time.
  • Fixed some bugs in HRP and HERC when we add constraints.
  • Fixed a bug in the duplication_summation_matrix.
  • Fixed tight layout in plot functions that uses multiple axes.

Version 6.2.0

  • Improvement in calculation speed of duplication_matrix, duplication_elimination_matrix and duplication_summation_matrix functions using a vectorized formula.
  • Fixed formulation of risk parity with risk factors model that produced incorrect results when using the MOSEK solver.
  • Fixed some bugs in PlotFunctions module.
  • Fixed some bugs in HCPortfolio related to custom_mu vector and use of Kurtosis and Semi Kurtosis as risk measures.
  • Standardized the way additional parameters to estimate mean vector and covariance matrix are entered.

Version 6.1.0

  • Implements standarized silhouette score to determine the optimal number of clusters.
  • Fix plot_clusters function to plot clusters and heatmap in same order of codependence matrix. Originally it plots the codependece matrix with axis x inverted.

Version 6.0.0

  • Implements risk parity optimization based on explicit risk factors and principal components.
  • Implements new formulations of Gini Mean Difference, Tail Gini, Range, CVaR Range and Tail Gini Range that improves speed compared to formulations based on the owa portfolio model.
  • Improves the calculation of elliptical uncertainty sets for worst case optimization.
  • Add new functions that allow us to calculate the risk contribution per explicit risk factors and principal components.
  • Add new functions that allow us to plot the risk contribution per explicit risk factors and principal components.

Version 5.0.0

  • Implements new kind of constraints that incorporates the information from networks like the Minimum Spanning Tree and Maximally Filtered Graph into the portfolio optimization models: return-risk portfolio, owa portfolio and worst case portfolio.
  • Implements new kind of constraints that incorporates the information from dendrograms into the portfolio optimization models: return-risk portfolio, owa portfolio and worst case portfolio.
  • Improves the speed of several functions using the c++ linear algebra library Eigen and c++ eigenvalues library Spectra.
  • Add new functions that allow us to plot the relationship between graphs and asset allocation.
  • Add new functions that allow us to create constraints based on graphs information.
  • Add a new example about applications of networks and dendrograms constraints in portfolio optimization problems.
  • Fixed some errors related to HCPortfolio with constraints.
  • Fixed some errors in some plots.

Version 4.4.0

  • Implements the approximate Kurtosis model through sum of squared quadratic forms for large scale kurtosis optimization.
  • Add the block vectorization operator.

Version 4.3.0

  • Implements custom constraints for the Relaxed Risk Parity portfolio model.
  • Add three new methods to estimate the mean vector: James-Stein, Bayes-Stein and BOP.

Version 4.2.0

  • Implements constraints for the Hierarchical Equal Risk Contribution (HERC) and Nested Clustered Optimization (NCO) portfolio models.
  • Add the option to show risk contributions as a percentage of total risk in risk contribution plot.
  • Repairs some bugs.

Version 4.1.0

  • Implements the Relativistic Value at Risk and Relativistic Drawdown at Risk portfolio models.
  • Implements the Higher L-moments portfolio model function as an special case of OWA portfolio.
  • Adds functions to calculate L-moments.
  • Adds a function to calculate risk contribution constraints on asset classes.
  • Repairs some bugs.

Version 4.0.0

  • Implements Kurtosis and Semi Kurtosis portfolio models based on parametric approach.
  • Implements new c++ based functions to speed up kurtosis model calculations.
  • Repairs some bugs.

Version 3.3.0

  • Adds Kendall Tau and Gerber statistic as options for codependence matrix in HCPortfolio object.
  • Adds Gerber statistic as an option for covariance matrix estimator in Portfolio and HCPortfolio objects.

Version 3.2.0

  • Implements reformulations of portfolio models based on drawdowns to speed up calculations.
  • Adds some tests for portfolio object and hcportfolio object.

Version 3.1.0

  • Implements a reformulation of OWA portfolio optimization to speed up calculations.

Version 3.0.0

  • Implements 5 additional risk measures for mean risk model: Gini Mean Difference, Tail Gini, Range, CVaR range and Tail Gini range.
  • Implements 4 additional risk measures for risk parity model: Gini Mean Difference, Tail Gini, CVaR range and Tail Gini range.
  • Implements the OWA Portfolio Optimization model for custom vector of weights and a module to build OWA weights for some special cases.
  • Implements a function to plot range risk measures.
  • Adds the option to use Graphical Lasso, j-Logo, denoising and detoning covariance estimates.

Version 2.0.0

  • Implement Nested Clustered Optimization (NCO) model with four objective functions.
  • Implements the Relaxed Risk Parity model.
  • Implements the Risk Budgeting approach for Risk Parity Portfolios with constraints.
  • Adds the option to use custom covariance in Hierarchical Clustering Portfolios.

Version 1.0.0

  • Redesigns of Riskfolio-Lib interface (Only import riskfolio for all functions).
  • Implements Hierarchical Risk Parity (HRP) model with constraints on assets' weights.
  • Implements a function that helps to build constraints for the HRP model.
  • Implements the Direct Bubble Hierarchical Tree (DBHT) linkage method for HRP and HERC models.
  • Implements a function that plots relationship among assets in a network using Minimum Spanning Tree (MST) and Planar Maximally Filtered Graph (PMFG).
  • Adds two new codependence measures: mutual information and lower tail dependence index.

Version 0.4.0

  • Implements Hierarchical Equal Risk Contribution with equally weights within clusters (HERC2).
  • Implements a function that help us to discretize portfolio weights into number of shares given an investment amount.
  • Implements the option to select the method to estimate covariance in HRP, HERC and HERC2.
  • Adds the option to add constraints on the number of assets and the number of effective assets.
  • Fixes an error in two_diff_gap_stat() when number of assets is too small.
  • Fixes an error on forward_regression() and backward_regression() when there is no significant feature in regression modes using p-value criterion.
  • Adds an example that shows how to build HERC2 portfolios.
  • Adds an example that shows how to build constraints on the number of assets and number of effective assets.

Version 0.3.0

  • Implements Hierarchical Risk Parity (HRP) and Hierarchical Equal Risk Parity (HERC).
  • Implements the function plot_clusters() and plot_dendrogram() that help us to identify clusters based on a distance correlation metric.
  • Implements the function assets_clusters() that help us to create asset classes based on hierarchical clusters.
  • Adds an example that shows how to build Hierarchical Risk Parity portfolios.
  • Adds an example that shows how to build Hierarchical Equal Risk Parity portfolios.

Version 0.2.0

  • Implements Logarithmic Mean Risk (Kelly Criterion) Portfolio Optimization models.
  • Implements the function plot_bar() that help us to plot portfolios with negative weights.
  • Adds the option to build dollar neutral portfolios.
  • Adds an example that shows how to build Logarithmic Mean Risk (Kelly Criterion) portfolios.
  • Adds an example that shows how to build dollar neutral portfolios.

Version 0.1.5

  • Adds the option to add a constraint on minimum portfolio return.
  • Adds an example of how to add constraints on portfolio return and risk measures.

Version 0.1.4

  • Adds Black Litterman with factors in two flavors: Black Litterman Bayesian model and Augmented Black Litterman model.
  • Implements factors_views, a function that allows to design views on risk factors for Black Litterman with factors.
  • Repairs some bugs.

Version 0.1.2

  • Adds Entropic Drawdown at Risk for Mean Risk Portfolio Optimization and Risk Parity Portfolio Optimization.
  • Repairs some bugs.

Version 0.1.1

  • Repairs some bugs in Portfolio related to Semi Variance and UCI.
  • Implements an option to annualize returns and risk in plot_frontier, Jupyter Notebook and Excel reports.
  • Adds examples using Vectorbt for Backtesting and MOSEK for large scale problems.

Version 0.1.0

  • Repairs some bugs in RiskFunctions.
  • Implements the Reports module that helps to build reports on Jupyter Notebook and Excel.
  • Implements plot_table, a function that resume some indicators of a portfolio.
  • Adds Entropic Value at Risk for Mean Risk Portfolio Optimization and Risk Parity Portfolio Optimization.

Version 0.0.7

  • Implements normal assumption method to estimate box and elliptical uncertainty sets for Worst Case Optimization.
  • Implements elliptical uncertainty sets for covariance matrix.
  • Adds Ulcer Index for Mean Risk Portfolio Optimization and Risk Parity Portfolio Optimization.
  • Implements functions to calculate Ulcer Index.

Version 0.0.6

  • Repairs some bugs.
  • Implements bootstrapping methods to estimate box and elliptical uncertainty sets for Worst Case Optimization.
  • Implements Worst Case Mean Variance Portfolio Optimization using box and elliptical uncertainty sets.

Version 0.0.5

  • Repairs some bugs.
  • Implements Risk Parity Portfolio Optimization for 7 convex risk measures.

Version 0.0.4

  • Repairs some bugs.
  • Update to make it compatible with cvxpy >=1.1.0
  • Implements Principal Component Regression for loadings matrix estimation.
  • Adds Akaike information criterion, Schwarz information criterion, R squared and adjusted R squared feature selection criterions in stepwise regression.

Version 0.0.3

  • Repairs some bugs.
  • Implements an option for building constraints common for all assets classes.

Version 0.0.2

  • Repairs some bugs.

Version 0.0.1

  • Implements robust and ewma estimates.
  • Implements Black Litterman model and risk factors models.
  • Implements mean risk optimization with 10 risk measures.