tomespel / kairos-volatility-prediction Goto Github PK

Advanced machine learning techniques for volatility prediction on financial markets. The Kairos package was developed for CFM’s ENS DataChallenge 2018.

Home Page: https://challengedata.ens.fr/en/challenge/34/volatility_prediction_in_financial_markets.html

License: Apache License 2.0

Python 0.22% Jupyter Notebook 99.78%

financial-analysis volatility machine-learning

kairos-volatility-prediction's Introduction

Volatility Prediction in Financial Markets

A repository for the Kairos package developed for CFM's ENS Data Challenge. Further information about this challenge is available at the following link. The official leaderboard is available at this address.

Background

Volatility is a fundamental metric in modern finance. In this challenge, we use past volatilities and price changes of financial instruments to predict future volatility and control the risk of financial portfolios. Here is the video of the challenge's presentation at Collège de France.

Kairos refers to an ancient Greek deity. The word καιρός itself means opportune moment. In modern greek, it means the weather.

Features

We have developed the Kairos package for predicting volatility at the end of the day. The package has been developed for the Python 3.7 language. Dependencies are specified in the package itself.

Installation

Usage

Loading the package

In order to use the Kairos package functions, you must import it in your Python 3.7 project.

from kairos import *

Make sure that all the dependencies are installed. When imported correctly, the package should print the following in the Python console.

Kairos features provider imported.
Kairos engine imported.
Kairos succesfully loaded in your project.

Input Files

The input files should be using the .csv format and have the following structure. We recommend using ',' as separator in the file.

ID	date	product_id	volatility 09:30:00	...	volatility 13:55:00	return 09:30:00	...	return 13:55:00
1	1	1	0.6627	...	0.0523	-1	...	-1
2	1	2	0.2853	...	0.0711	1	...	-1
3	1	3	1.1516	...	0.1179	-1	...	1

In order to import a csv file in Kairos, use the following command.

myData = kairos.input.from_csv('/csvFilePath/myFile.csv')

You can specify the csv delimiter by adding the parameter contentsDelimiter in the function.

File system

Kairos uses a dedicated file system .krf to store the data in their current state. In order to store the data, use the following command.

kairos.output.freeze(myData, '/pathToFile/myFile.krf')

You can restore the data later on using the following command.

myData = kairos.input.release('/pathToFile/myFile.krf')

The data is structured in one or several KairosBucket objects which include KairosAsset objects with similar properties. Each KairosAsset includes several KairosDay objects.

Credits

The team includes the following members, with the associated GitHub identifiers.

Name	GitHub id
Mehdi Tomas	mehditomas
Thomas Espel	tjespel

License

Apache 2

kairos-volatility-prediction's People

Contributors

Watchers

kairos-volatility-prediction's Issues

Lay out main structure

Lay out the main file structure.

Main task	Primal Component	Secondary Component
Asset Classifier	High Average Volatility	High Local Volatility
		Medium Local Volatility
		Low Local Volatility
	Medium Average Volatility	High Local Volatility
		Medium Local Volatility
		Low Local Volatility
	Low Average Volatility	High Local Volatility
		Medium Local Volatility
		Low Local Volatility
Asset Extractor	Main Asset Extractor	n/a: Main Component
	High Local Volatility	Raw Data Processing
	Medium Local Volatility	Raw Data Processing
	Local Local Volatility	Raw Data Processing
Features Engineering	High Local Volatility	Features Engineering Provider
		Features Engineering Selector
	Medium Local Volatility	Features Engineering Provider
		Features Engineering Selector
	Low Local Volatility	Features Engineering Provider
		Features Engineering Selector
Predictive Engine	High Local Volatility	Features Engineering Collector
		Training Network
	Medium Local Volatility	Features Engineering Collector
		Training Network
	Low Local Volatility	Features Engineering Collector
		Training Network
Output Processor	High Local Volatility	Predictions Collector
	Medium Local Volatility	Predictions Collector
	Low Local Volatility	Predictions Collector
	Main Output Delivery	n/a: Main Component

Define the list of elements in the KairosDay class

Following #4 . Here are the elements of the class KairosDay.

__id (int) corresponds to the target id.
_asset (int) corresponds to the asset id.
__date (int) corresponds to the date.
__volatility corresponds to the list of double with the volatility values.
__returns corresponds to the list of double with the returns values.
__isClassified (bool) indicates if the asset has received it local volatility classification.
__localClassification (string) specifies the local classification hlv, mlv or llv. States None otherwise.
__classificationDetails (dict) specifies the local classification probabilities in the form of a dictionary {'method': double(probability)}.
__features (dict) specifies the features list in the form of a dictionary {'featureName': list(featureValuesList)}.

What do you think?