https://github.com/axsaucedo/sparse

The effects of sparse and group-feature regression models in portfolio optimization.
https://github.com/axsaucedo/sparse

Last synced: about 7 hours ago
JSON representation

The effects of sparse and group-feature regression models in portfolio optimization.

• Host: GitHub
• URL: https://github.com/axsaucedo/sparse
• Owner: axsaucedo
• License: mit
• Created: 2014-04-19T19:07:49.000Z (over 10 years ago)
• Default Branch: master
• Last Pushed: 2014-05-18T09:49:40.000Z (over 10 years ago)
• Last Synced: 2024-04-15T03:22:16.904Z (7 months ago)
• Language: Matlab
• Homepage:
• Size: 644 MB
• Stars: 22
• Watchers: 5
• Forks: 10
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.txt

Awesome Lists containing this project

README

        
# Sparse and Group Regression models in Portfolio Optimization 

## Introduction

This repo contains the implementation of models studied, analysed and proposed in ["The effects of Sparse and Group Regression models in Portfolio Optimization"](../master/paper.pdf?raw=true).

This implementation focuses on finding the effects of Sparse and Group regression approaches to portfolio optimization problems in finance.

## Motivation

Current approaches to portfolio optimization consider stocks as individual entities, and do not exploit the grouping/classifying information available (e.g. Financial Sectors, Industries, Type, etc).

This paper proposes a novel approach to Index Tracking - namely, a sparse, group and sparse group approach.

## Implementation

This repo contains the implementation of the following models:

#### Feature Regression Models

* Absolute Values

* Conditional-Value-at-Risk (CVaR) Optimization

* Norm-Constrained CVaR Optimization

* Lasso

#### Group Regression Models

* Group Selection

* Group Lasso

* Sparse Group Lasso

## Requirements

This implementation requires the [CVX library](http://cvxr.com/cvx/download/) for solving the convex optimization problems.

## Usage

Tests were built to provide intuition when implementing the Sparse Group Regression model into a set of data, however, understanding on these models is required for an effective use;