/ML_Prediction_TaLib

machine learning prediction algorithm for stock returns (in progress)

Primary LanguageJupyter Notebook

ML Stock Market Prediction Algorithm using TaLib Library

Description

The model is trained on the basis of positive/negative daily returns. It is trained by using the 60 technical idicators provided in the TaLib Library.

Prerequisites

Requirements of libraries

  • TaLib
  • Scikit Learn

Usage

So far, I have usued all technical indicators provided by the library, this is of course not the best solution. But as I continue the project, I will optimise which technical indicators I use. With the current set, the accuracy of the prediction varies between 78-85%, depending on the chosen stock.

Strategy Returns for Google (GOOG)

Returns

Accuracy Report

Accuracy Report

Roadmap

  • Find best set of Technical Indicators to maximize accuracy