Webinar Recap: Leveraging Insights in Data for 2021 Bull Run & the Role of Machine Learning



A panel of FinTech experts from Singapore shares how to leverage data insights through machine learning to generate financial alphas.


Data is the new oil in the age of digital economy.

On 28 Jan 2021 (Thu), Smart Solutions brings you a 90-min online panel with invited finance and Fintech experts to share their perspectives on leveraging technical and fundamental data and its trending insights for financial strategy building and modelling to generate alphas for 2021.


Hosted by Genevieve Goh (CEO, Smart Solutions) in collaboration with Tejas Padnis (Senior Data Scientist, Smart Solutions), and Bao Vu (Investment Director, R.E. Lee Capital)


Here are the topics you can expect:

1. Harnessing Trends Through Advanced Financial Data Science

Showcase of how to apply co-relevancy of data from raw financial data extracted. How to extract trends from it and using it to help decision making.

Tejas Padnis - Senior Data Scientist, Smart Solutions (Smarts.sg)

2. Generating An Impressive Market-Beating Alpha Run in 2020 Through Machine Learning

Sharing on how Bao maintained an alpha return in an eventful year by using financial machine learning.

Bao Vu - Investment Director, RE Lee Capital


Webinar Timestamp

Harnessing Trends through advanced Financial Data Science

0:53 - SMP500 Projections

1:20 - Where is Leveraging AI & ML for Alpha in Cumulative Forecasted Revenue from 2016 - 2023

3:44 - Financial Data Science Project Lifecycle

7:35 - Data Collection

9:09 - Feature Engineering

12:00 - Feature selection

15:43 - Machine Learning Models

18:43 - Backtests

23:10 - Forecast/Prediction


Generating An Impressive Market Beating Alpha in 2020 through ML

1:48 - Firm (R.E. Lee Capital) background

2:40 - Bao Vu’s background

4:43 - Why you should care about integrating Machine Learning into Investment Process

6:25 - Various Investment Approaches

8:28 - R.E. Lee Capital’s chosen Approach - Beta

10:00 - Background on ML portfolio & mandate

11:24 - Defining the Prediction Problem in Machine Learning

13:30 - Feature Engineering for R.E. Lee Capital

14:40 - Machine Learning Data Inputs (Macro)

17:20 - Machine Learning Data Inputs (Specific Index)

19:38 - Feature Selection

20:45 - Beta Model Forecast Prediction

22:12 - Results they had

25:00 - Q&A



Let’s take a closer look at some of the questions and key insights from the discussion:


Are There Any Expected Increase in Revenue for Investment Companies Using AI?


According to a report by Statista, algorithm trading strategies and performance improvement is ranked at number two in terms of the cumulative forecasted revenue that is going to be generated from 2016 to 2023. This adds volume to AI-backed investment strategies and justifies its relevance in today’s growing market.


As predicted in the market, Bull Run is scaling and expected to grow. And if we go by the statistics, it can produce massive revenue for the investment companies that are using algorithm-based trading strategies to boost their existing mechanisms.


As explained by Goh, “Leveraging data can help you rise with the Bull Run of 2021 considering the above information provided by reliable sources.’


Do Algorithms Depend Heavily on Data, or is the Truth Something Else?


Machine Learning helps in maneuvering accurate investment predictions through algorithm-based trading. At the foundational level, data plays a huge role in howMachine Learning and Artificial Intelligence can be effectively leveraged for generating expected returns to organizations.


The part that amazes us is how these algorithms are using data to create impactful and powerful investment strategies and redefining goals for all the stakeholders of the industry.


How does a Life Cycle of a Data Science Project Impacts the Machine Learning Model for Your Finance Business?


Tejas Padnis believes that the Machine Learning model has some powerful insights that prove the role of data harnessing as a critical part of the life cycle of any data science project. Even in an investing model, Machine Learning algorithms are influenced by the type of database input provided.


The data is gathered from various resources such as financial markets, news, and social media in the form of structured or unstructured data. It is then filtered through the feature engineering process. The importance of feature engineering is to make the data set used for machine learning. It helps sort the database with inputs and make it compatible for improving the performance of machine learning models.


Why Is Quality Data Important for Machine Learning?


Research, Analytics, and Forecast are the three stages of the data which is the lifeblood of the project. But the whole process revolves around gathering data and preparing it. One has to be sure of different aspects related to data right from the quality, the volume, to the accuracy. Also, data should be capable of solving the problem that you are working on. Hence, data harnessing becomes relevant for machine learning.


Why Is Backtesting the Most Important Part of the Financial Data Life Cycle Project?


As shared by Tejas, backtesting refers to the use of trading strategies based on historical data, and the estimates of the past data for performance. It compares different strategies for traders to produce successful results through an Anchored walk forward and Non-anchored forward mechanisms. Back-testing is a model reliable for producing powerful direction to machine learning algorithms that further enhances the investment processes in the companies.


What Role Is RoboInvestor Playing?


Bao Vu highlighted the application of RoboInvestor by Smart Solution as a Machine Learning backed software to strengthen investment and stock trading practices. His company, R.E. Lee Capital is actively using RoboInvestor, focusing on the ‘Beta’ model for Machine Learning to maneuver investment portfolios to generate reliable predictions with hefty accuracy.


One of the core offerings by Smart Solutions, RoboInvestor leverages machine learning algorithms to tailor solutions for more accurate investment management processes.


Bao emphasizes that while working with Machine Learning, you must know about the prediction expectations and the problems you want to solve. What is the focus? Is it index, stock prices or something else? It is extremely important to define what the model is sort of predicting.


How Machine Learning Has Helped R.E. Lee Capital in Scaling Their Business Revenues?


In the webinar, Bao Vu has extensively explained the whole process and the role of RoboInvestor in giving direction to the organization. With some rock-solid insights and elucidated relevance, it has made trading strategies forecast more accurate and dependable. Also, discover how in 2020, the company was able to produce value, growth, momentum, and more in the factor/stock selection process.

Machine Learning, if backed by well-featured data can be a powerful source for investment companies to receive an unprecedented return in the future. The time has come when the realization of adapting to Machine Learning should be considered vital by the finance market.


Watch the complete webinar for better insights and understanding on the topic:

Part 1


Part 2


0 comments