What is the MIG Platform
This page will go over what the MIG platform is, it's purpose, and where it's heading.
Motivations
The MIG platform was conceived in W21 on the quant and investment side, seeking to bring the two sides of the club together. Traditionally, both sides were focused on their own separate ideas and this slowly devolved into sort of "two clubs into one". As a result, the quant team focused more on systematic trading (but with no money), and the investment side focused on managing the traditional fund (amounting to around $6,000 as of this writing). The Platform would enable MIG to become a club that focused not only on investment but tech-empowered investing.
The Features
Portfolio Tracking
One of the issues the initial team found was the fact that there was no way to accurately track the portfolio for the club. This often resulted in large amounts of confusion on how specific stocks were allocated to the portfolio. In order to promote more transparency and commitment, the MIG platform will be the first platform to actively track a club's portfolio dynamically.
Theses Aggregations
As an investment member, identifying key theses, assumptions, and catalysts for various stocks is an important characteristic for determining investment. One of the main issues in investment meetings is the lack of dissemination of thought post-presentation. The platform enables users to add fair value estimations and write thesis statements (both bullish and bearish) that can later be used to make better fair value predictions along with better reasons to invest in a stock.
Fair Value - Wisdom of the Crowd
Fair value calculations have traditionally been done by lofty assumptions given by bias. Having everyone produce a fair value calculation given a presentation or research will ultimately lead to more accurate "fair value" estimation for various stocks, attributing to the "Wisdom of the Crowd" mentality. This is built into the platform.
Earnings Call Analysis
Earnings calls are long, tedious, and are often positively sentiment. There needs to be a better way of analyzing earnings calls that takes out the "bias" and focuses on the meaningful content that pertains to innovation and the ability for the company to truly innovate (confidence score). Our earnings call analysis algorithms that utilize FinBERT, and TARS-Classifier for Semi-Supervised Learning allow for better labeling of sentences that are aggregated to produce a final confidence score.
Sentiment Over Time
Identifying the company sentiment over time is an extremely useful task that allows us to quickly gauge whether or not the company is gaining or losing favor with investors and the crowd (via news articles crawled from multiple source). Using this information, investors can also pinpoint sentiment as a potential opportunity.
Conclusion
The platform is an ever-evolving, constantly-changing project and we're super excited to see where it continues to move forward. Please pay attention to the various statuses of the current project and if you have any suggestions or ideas, please reach out to the current platform lead.
Last updated
