An algorithmic trading bot that learns and adapts to new data and evolving markets using Financial Python Programming and Machine Learning.

Overview

ALgorithmic_Trading_with_ML

An algorithmic trading bot that learns and adapts to new data and evolving markets using Financial Python Programming and Machine Learning.

The following steps are followed :

  • Establishing a Baseline Performance
  • Tuning the Baseline Trading Algorithm
  • Evaluating a New Machine Learning Classifier
  • Creating an Evaluation Report

Establishing a Baseline Performance

  1. Importing the OHLCV dataset into a Pandas DataFrame.

  2. Trading signals are created using short- and long-window SMA values.

svm_original_report

  1. The data is splitted into training and testing datasets.

  2. Using the SVC classifier model from SKLearn's support vector machine (SVM) learning method to fit the training data and making predictions based on the testing data. Reviewing the predictions.

  3. Reviewing the classification report associated with the SVC model predictions.

svm_strategy_returns

  1. Creating a predictions DataFrame that contains columns for “Predicted” values, “Actual Returns”, and “Strategy Returns”.

  2. Creating a cumulative return plot that shows the actual returns vs. the strategy returns. Save a PNG image of this plot. This will serve as a baseline against which to compare the effects of tuning the trading algorithm.

Actual_Returns_Vs_SVM_Original_Returns


Tune the Baseline Trading Algorithm

The model’s input features are tuned to find the parameters that result in the best trading outcomes. The cumulative products of the strategy returns are compared. Below steps are followed:

  1. The training algorithm is tuned by adjusting the size of the training dataset. To do so, slice your data into different periods.

10_month_svm_report 24_month_sw_4_lw_100_report 48month_sw_4_lw_100_report

Answer the following question: What impact resulted from increasing or decreasing the training window?

Increasing the training dataset size alone did not improve the returns prediction. The precision and recall values for class -1 improved with increase in training set data and presion and recall values for class 1 decreased compared to the original training daatset size(3 months)

  1. The trading algorithm is tuned by adjusting the SMA input features. Adjusting one or both of the windows for the algorithm.

Answer the following question: What impact resulted from increasing or decreasing either or both of the SMA windows?

  • Increasing the short window for SMA increased impacted the precision and recall scores. It improves these scores till certain limit and then the scores decreases.
  • While increasing the short window when we equally incresase the long window we could achieve optimal maximized scores.
  • Another interesting obervation is that when the training dataset increses the short window and long window has to be incresed to get maximum output.

3_month_sw_8_lw_100_report

The set of parameters that best improved the trading algorithm returns. 48_month_sw_10_lw_270_report 48_month_sw_10_lw_270_return_comparison


Evaluating a New Machine Learning Classifier

The original parameters are applied to a second machine learning model to find its performance. To do so, below steps are followed:

  1. Importing a new classifier, we chose LogisticRegression as our new classifier.

  2. Using the original training data we fit the Logistic regression model.

  3. The Logistic Regression model is backtested to evaluate its performance.

Answer the following questions: Did this new model perform better or worse than the provided baseline model? Did this new model perform better or worse than your tuned trading algorithm?

This new model performed good but not as well as our provided baseline model or the tuned trading algorithm.

lr_report lr_return_comparison

Graph InfoClust: Leveraging cluster-level node information for unsupervised graph representation learning

Graph-InfoClust-GIC [PAKDD 2021] PAKDD'21 version Graph InfoClust: Maximizing Coarse-Grain Mutual Information in Graphs Preprint version Graph InfoClu

Costas Mavromatis 21 Dec 03, 2022
Zero-shot Learning by Generating Task-specific Adapters

Code for "Zero-shot Learning by Generating Task-specific Adapters" This is the repository containing code for "Zero-shot Learning by Generating Task-s

INK Lab @ USC 11 Dec 17, 2021
A Python 3 package for state-of-the-art statistical dimension reduction methods

direpack: a Python 3 library for state-of-the-art statistical dimension reduction techniques This package delivers a scikit-learn compatible Python 3

Sven Serneels 32 Dec 14, 2022
Trajectory Variational Autoencder baseline for Multi-Agent Behavior challenge 2022

MABe_2022_TVAE: a Trajectory Variational Autoencoder baseline for the 2022 Multi-Agent Behavior challenge This repository contains jupyter notebooks t

Andrew Ulmer 15 Nov 08, 2022
REBEL: Relation Extraction By End-to-end Language generation

REBEL: Relation Extraction By End-to-end Language generation This is the repository for the Findings of EMNLP 2021 paper REBEL: Relation Extraction By

Babelscape 222 Jan 06, 2023
Multitask Learning Strengthens Adversarial Robustness

Multitask Learning Strengthens Adversarial Robustness

Columbia University 15 Jun 10, 2022
A PyTorch implementation for V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation

A PyTorch implementation of V-Net Vnet is a PyTorch implementation of the paper V-Net: Fully Convolutional Neural Networks for Volumetric Medical Imag

Matthew Macy 606 Dec 21, 2022
Breast Cancer Classification Model is applied on a different dataset

Breast Cancer Classification Model is applied on a different dataset

1 Feb 04, 2022
Trained on Simulated Data, Tested in the Real World

Trained on Simulated Data, Tested in the Real World

livox 43 Nov 18, 2022
Implementation of Lie Transformer, Equivariant Self-Attention, in Pytorch

Lie Transformer - Pytorch (wip) Implementation of Lie Transformer, Equivariant Self-Attention, in Pytorch. Only the SE3 version will be present in thi

Phil Wang 78 Oct 26, 2022
PRIME: A Few Primitives Can Boost Robustness to Common Corruptions

PRIME: A Few Primitives Can Boost Robustness to Common Corruptions This is the official repository of PRIME, the data agumentation method introduced i

Apostolos Modas 34 Oct 30, 2022
AEI: Actors-Environment Interaction with Adaptive Attention for Temporal Action Proposals Generation

AEI: Actors-Environment Interaction with Adaptive Attention for Temporal Action Proposals Generation A pytorch-version implementation codes of paper:

11 Dec 13, 2022
Source code for our EMNLP'21 paper 《Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning》

Child-Tuning Source code for EMNLP 2021 Long paper: Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning. 1. Environ

46 Dec 12, 2022
Pytorch implementation of "ARM: Any-Time Super-Resolution Method"

ARM-Net Dependencies Python 3.6 Pytorch 1.7 Results Train Data preprocessing cd data_scripts python extract_subimages_test.py python data_augmentation

Bohong Chen 55 Nov 24, 2022
AlgoVision - A Framework for Differentiable Algorithms and Algorithmic Supervision

NeurIPS 2021 Paper "Learning with Algorithmic Supervision via Continuous Relaxations"

Felix Petersen 76 Jan 01, 2023
LowRankModels.jl is a julia package for modeling and fitting generalized low rank models.

LowRankModels.jl LowRankModels.jl is a Julia package for modeling and fitting generalized low rank models (GLRMs). GLRMs model a data array by a low r

Madeleine Udell 183 Dec 17, 2022
Aalto-cs-msc-theses - Listing of M.Sc. Theses of the Department of Computer Science at Aalto University

Aalto-CS-MSc-Theses Listing of M.Sc. Theses of the Department of Computer Scienc

Jorma Laaksonen 3 Jan 27, 2022
Minimal PyTorch implementation of YOLOv3

A minimal PyTorch implementation of YOLOv3, with support for training, inference and evaluation.

Erik Linder-Norén 6.9k Dec 29, 2022
Re-implement CycleGAN in Tensorlayer

CycleGAN_Tensorlayer Re-implement CycleGAN in TensorLayer Original CycleGAN Improved CycleGAN with resize-convolution Prerequisites: TensorLayer Tenso

89 Aug 15, 2022
Baleen: Robust Multi-Hop Reasoning at Scale via Condensed Retrieval (NeurIPS'21)

Baleen Baleen is a state-of-the-art model for multi-hop reasoning, enabling scalable multi-hop search over massive collections for knowledge-intensive

Stanford Future Data Systems 22 Dec 05, 2022