Lazy Trading Part 7: Developing self-adapting Trading System

Learn to assemble Smart Self Learning Algorithms. Predict future price change based on financial data patterns

4.05 (24 reviews)
Udemy
platform
English
language
Investing & Trading
category
473
students
4.5 hours
content
Jun 2021
last update
$49.99
regular price

What you will learn

Log data from financial assets to files

Learn to use Deep Learning with H2O

Setup Automated Decision Support Loop

Automate R scripts

Develop R code

Use Version control for your R project

Writing R functions

Perform data manipulations with pipes

Use H2O Machine Learning platform in R

Perform Deep Learning on Time-Series data

Evaluate performance of Deep Learning models

Backtest trading strategy in R Software

Description

"No one can promise that this will work, at least it will work by itself!"

About the Lazy Trading Courses:
This series of courses is designed to to combine fascinating experience of Algorithmic Trading and at the same time to learn Computer and Data Science! Particular focus is made on building Decision Support System that can help to automate a lot of boring processes related to Trading and also learn Data Science. Several algorithms will be built by performing basic data cycle 'data input-data manipulation - analysis -output'. Provided examples throughout all 7 courses will show how to build very comprehensive system capable to automatically evolve without much manual input.

About this Course: Developing Self Learning Trading Robot with Statistical Modeling

This course will cover usage of Deep Learning Regression Model to predict future prices of financial asset. This course will blend everything that was previously explained to use:

  • Use MQL4 DataWriter robot to gather financial asset data

  • Use R Statistical Software to aggregate data to be ready for modeling

  • Use H2O Machine Learning Platform to train Deep Learning Regression Models

    • Use random neural network structures

    • Functions with test and examples in R package

  • Back-test trading strategy using Model prediction and historical data

  • ... update model if needed

  • Use Model and New Data to generate predictions

  • Use Model output in MQL4 Trading Robot

  • Adding and using Market Type info [from course 6]

  • Experiment by adding Reinforcement Learning to select best possible Market Type

  • Try easy to deploy ready to use complex Trading System

"What is that ONE thing very special about this course?"

-- Watch AI predicting the future!

This project is containing several courses focused to help managing Automated Trading Systems:

  1. Set up your Home Trading Environment

  2. Set up your Trading Strategy Robot

  3. Set up your automated Trading Journal

  4. Statistical Automated Trading Control

  5. Reading News and Sentiment Analysis

  6. Using Artificial Intelligence to detect market status

  7. Building an AI trading system

IMPORTANT: all courses will have a 'quick to deploy' sections as well as sections containing theoretical explanations.

What will you learn apart of trading:

While completing these courses you will learn much more rather than just trading by using provided examples:

  • Learn and practice to use Decision Support System

  • Be organized and systematic using Version Control and Automated Statistical Analysis

  • Learn using R to read, manipulate data and perform Machine Learning including Deep Learning

  • Learn and practice Data Visualization

  • Learn sentiment analysis and web scrapping

  • Learn Shiny to deploy any data project in hours

  • Get productivity hacks

  • Learn to automate your tasks and scheduling them

  • Get expandable examples of MQL4 and R code

What these courses are not:

  • These courses will not teach and explain specific programming concepts in details

  • These courses are not meant to teach basics of Data Science or Trading

  • There is no guarantee on bug free programming

Disclaimer:

Trading is a risk. This course must not be intended as a financial advice or service. Past results are not guaranteed for the future. Significant time investment may be required to reproduce proposed methods and concepts

Content

Introduction

Specific Goals for this Course
Disclaimer

How to predict the future with the past data? Getting to know the idea!

Refresher - Decision Support System
Types of Algo-Trading systems [Author's view]
Optimization vs Pattern Recognition
Predict the future with Deep Learning - Intuition

Get the Code

How to get the code?
Assemble the puzzle - tasks overview
R package 'lazytrade'

Predict the future with Deep Learning - Code

Load data function
Transposed Data Function
Labeled Data Function
Join, Shift, Split data to Training and Test Datasets
Deep Learning Classification and Regression models
Strategy Back-test Function
[update]Trust but check - Logging Strategy Test results to check how AI performs
Trying different options
Self Learn Script - Re-train Deep Learning Models
Score Data Script - Generate Future Predictions
Log prediction results to the file

Scheduling Decision Support System

Overview of Scheduled Tasks
Self-Learning Tasks
Predicting Tasks

Trading Robot FALCON_F2: Live Demo Testing, Debugging, etc

Important Notice: Work In Progress
Reading Predictions from Decision Support System
FALCON_F - Overview
FALCON_F - Working example
Trading Robot FALCON_F2 - Functions are better!
Creating the Dashboard

Trading Robot Falcon A - Model based trading system

Introducing another model based trading robot that learns by itself
FALCON_A Get the code
FALCON_A - reverse engineering
k-Nearest Neighbors intuition
k-NN using R
FALCON_A Code Overview
FALCON_A Optimization/Training
Summary: FALCON_A Robot - Learning from successes and mistakes

Demo Trading Experiment with FALCON_F2 and FALCON_A

Goals of this Demo Trading Experiment
Week1 results
Week2 results
Week3 results
Week4 results

Price change predictive Model for every Currency Pair

Motivation of this section

Conclusion for Part 7

Summary of this course
Your special *BONUS*

Bonus Section

Lazy Trading Overview

Screenshots

Lazy Trading Part 7: Developing self-adapting Trading System - Screenshot_01Lazy Trading Part 7: Developing self-adapting Trading System - Screenshot_02Lazy Trading Part 7: Developing self-adapting Trading System - Screenshot_03Lazy Trading Part 7: Developing self-adapting Trading System - Screenshot_04

Reviews

Michael
May 18, 2021
The author knows R language more than A-Z and, of course, MQL4. It is not difficult to understand him. Here in this course, I discovered so many new things. It covers - IT, programming (MQL4 and R), forecasts, automatization cycles. If someone is new I suggest buying all series, where it is all explained, and depicted.
Florian
December 13, 2020
All the Lazy trading parts was awesome! I did learn a lot, not only for trading but also some other computer science useful at work or for myself. I have now a very nice system for trading that I could implement with my own robots to improve my trading quality. The part 7 bring a new approach of the deep learning and get the MQL4 script using it. This series is brilliant, the instructor is very active and answers to all questions. I made a good choice to participate to the lazy trading course.
Surapoom
January 4, 2019
The idea of integrating deep learning to help optimizing trading strategy is very sound. I have learnt greatly. Thanks.
Roland
October 20, 2018
excellent way to learn deep learning. not sure if this method will work for this trading purpose however this method can be certainly applicable to other areas of AI

Charts

Price

Lazy Trading Part 7: Developing self-adapting Trading System - Price chart

Rating

Lazy Trading Part 7: Developing self-adapting Trading System - Ratings chart

Enrollment distribution

Lazy Trading Part 7: Developing self-adapting Trading System - Distribution chart
1482480
udemy ID
12/26/2017
course created date
11/22/2019
course indexed date
Bot
course submited by