A one-day course introducing you to Alternative Data and Data Science in the Trading and Investment world. The workshop will offer a combination of hands on exercises and a business overview in this emerging space. We will arm you with tools to build sample trading strategies using alternative data sets, data science and machine learning approaches for you to illustrate and generate positive PnL in financial markets.

Course Outline

Morning Session - Alternative Data and Data Science

---------

Alternative Data

-What is Alternative Data - an Overview of Alternative Data in Finance

-Categories of Alternative Data:

• web traffic, social and sentiment, and app usage
• credit/debit card transaction data
• web email and consumer receipts
• Geo-location, satellite, and weather
• sensor / IoT data

Data Science and Machine Learning Models for Finance

-Data cleansing and preparation

• Using Pandas to filter and clean data
• Aggregating and encoding categorical data into features

-Model Selection - Classical Models

• Simple features and shallow ML using linear and logistic regression with L1 and L2 regularizations.
• Features engineering and features selection

-Model Selection - Machine Learning Models

• Supervised vs unsupervised learning approaches
• Random forests and gradient boosted trees with Scikit-Learn and LightGBM
• Convolution Neural Networks (CNN) and Reinforcement Learning (RL)
• AutoML - open source solutions (Mindsdb, TPOT)
• AutoML - vendor solutions (Azure ML, Google AutoML, AWS Sagemaker, H20)

-Model Selection - goodness of fit and learning outcome metrics

-Backtesting - proving model performance

-Automated execution - choosing the best platform for your strategy

Afternoon Session - Hands on Exercises

---------

Environment Setup

-Environment Setup - Azure Notebooks, Google Colabs and Jupyter

-Environment Setup - Azure ML, Google AutoML and MindsDB

Sample Trading Strategies

-Equities - quarterly earnings surprise and disappointment (quarterly data)

• Online retailer example - ASOS (social media, web traffic)
• Online booking service example - Booking.com (social media, web traffic)
• Gaming example - Electronic Arts (public gaming stats API and Twitch streams)

-Equities - intra-quarterly trading strategy (daily data)

• Online booking service example - Booking.com (social media, web traffic)

Course Wrap up and Q&A

……………………………………..

Course Outcomes

The course participants are expected to gain hands on knowledge of data science modelling skills, trading strategies and insights into new unique data sets used by quantitative hedge funds today globally.

Audience

The course is intended for financial professionals, data scientists, research analysts and portfolio managers looking to get more exposure to the emerging alternative data space in finance.

Pre-requisites

An interest in alternative data in finance and data science. A basic understanding of any popular programming language such as Python is helpful but not required.

……………………………………..

About DataScrum:

DataScrum is a community of data scientists who organize fun and educational events in the alternative data space. You can keep up to date with their events and workshops by signing up to their weekly newsletter here and joining their meetup group. If you are looking for your next job opportunity reach out to them at team@datascrum.co.uk

……………………………………..

Instructor bio

Julian Kaljuvee is a co-founder of DataScrum, a London-based data science community.He is currently a data science consultant with Morningstar, a global financial data and analytics firm. Julian has over twenty years of experience in building quantitative data science and trading models for top financial institutions including Goldman Sachs, Morgan Stanley, JPMorgan, UBS, Barclays, HSBC and London Stock Exchange. Julian completed his bachelor’s degree in applied mathematics from Harvard University, Cambridge MA and pursued masters studies in statistics and probability theory at Columbia University, New York, NY.