Skip to content

Kevin20250000000/Quant-Trading-ML-Using-Support-Vector-Machine-Random-Forest-Neural-Network

Repository files navigation

Quant Trading - ML Using Support Vector Machine & Random Forest & and Neural Network

Strategy Overview:

  1. Pair the 30 stocks in the Dow Jones Industrial Average and identify the 10 pairs with the highest cointegration (i.e., the smallest p-value) as investment targets.

  2. Train a model using the Support Vector Machine (SVM) method and calculate the annualized return.

  3. Train a model using the Random Forest (RF) method and calculate the annualized return.

  4. Train a model using the Neural Network (NN) method and calculate the annualized return.

The average annual return from these three methods is then calculated for comparison.

Data:

DJI

Top 10 cointegrated pairs

Strategies:

I. Support Vector Machine (SVM)

SVM - Code

1. Top 10 Most Significant Cointegrated Stock Pair Spreads:

image

2. The Heatmap of Top 10 Strongest Cointegrated Pairs

image

3. The Mean Annualized Return of the SVM Model for the Top 10 Pairs is 59.02%.

image

4. Cumulative Returns of SVM Strategy for Top 10 Pairs:

image

II. Random Forest (RF)

RF - Code

The Mean Annualized Return of the Random Forest Model for the Top 10 Pairs is 59.54%.

image

III. Neural Network (NN)

NN - Code

The Mean Annualized Return of the Neural Networkt Model for the Top 10 Pairs is 41.77%.

image

Releases

No releases published

Packages

No packages published