MarketMoodRing🎭

Description

MarketMoodRing🎭 is a Python package designed for testing different regime detection models and portfolio optimizers. This tool is a product of research conducted by the UC Berkeley, Haas School of Business, Master of Financial Engineering, 2023. It aims to provide a framework for financial market regime analysis and portfolio management testing.

Collaborators

• Yves D'hondt
• Matteo Di Venti
• Rohan Rishi
• Jackson Walker

Reference

The \reference folder contains the project paper from where this library originated.

Features

• Regime Detection Models: Currently includes Hidden Markov Models (HMM) and Wasserstein K-Means clustering (WKM).
• Portfolio Optimizers: Implements different portfolio optimization strategies, including stochastic programming and factor-based optimization.

Installation

As of now, the MarketMoodRing package is not available on PyPI. To install and use this package, you need to clone the repository and reference it locally. Here are the steps to do so:

1. Clone the repository:

bash git clone https://github.com/yvesdhondt/MarketMoodRing.git

1. Navigate to the cloned directory:

bash cd MarketMoodRing

1. Now, you can import and use the package in your Python scripts. Make sure your script is in the same directory as the cloned repository or adjust the Python path accordingly.

```python import sys sys.path.insert(0, '/path/to/MarketMoodRing')

from marketmoodring.regime_detection import HiddenMarkovRegimeDetection ```

Please replace /path/to/MarketMoodRing with the actual path to the cloned repository on your system.

Remember to keep the repository updated with:

bash git pull origin main

Dependencies

The MarketMoodRing package requires several dependencies to function properly. These dependencies can be installed using either conda (recommended) or pip.

Using Conda

If you're using Conda, you can create a new environment and install all dependencies using the environment.yml file located in the root directory. Run the following command in your terminal / Anaconda Prompt once you've navigated to the cloned repository:

bash conda env create -f environment.yml

This will create a new Conda environment called marketmoodring-env and install all necessary packages. To activate the environment, use:

bash conda activate marketmoodring-env

Using pip

If you prefer using pip, you can install all dependencies using the requirements.txt file also located in the root directory. Run the following command in your terminal:

bash pip install -r requirements.txt

This will install all the necessary packages listed in the requirements.txt file. Please note that this will install the packages globally on your system, which can result in unexpected behavior. If you want to install the packages in a virtual environment, please refer to the Python documentation, or use the Conda environment as described above.

Usage

```python from marketmoodring.regimedetection import HiddenMarkovRegimeDetection from marketmoodring.portfoliooptimization import JointStochasticProgOptimization

Read in your data

indexdata = pd.readcsv('path/to/index_data.csv')

Initialize regime detection model

hmmmodel = HiddenMarkovRegimeDetection(nregimes=2, hmmtype='GMMHMM', covartype="diag", n_iter=100)

Fit the model to your data and predict regimes

fittedstates, fittedstatesproba = regimemodel.fittransform(indexdata)

Initialize portfolio optimizer

optmodel = JointStochasticProgOptimization(nregimes=2, objective="maxavgsharpe")

Fit the optimizer to your data and regime predictions and calculate portfolio weights

weights = optmodel.calculateweights( fittedstates = fittedstates, transmat = hmmmodel.gettransmat() indexdata = indexdata, ) ```

License

This project is licensed under the MIT License. See LICENSE for more details.