Technology and Innovation Community

A recent guest posting on Data Driven Venture by Ali Almufti, MD at BlackRock leading BlackRock's Systematic Private Investment Strategies, flagged a paper by Almufti and colleagues:

How Data-Driven Can Venture Capital Investing Really Be?

BlackRock JPM50th Systematic Insights into Private Equity Investing

It was nice to read how the authors test more rigorously the intuition that systematic investing could be applied to private equity assets. And why now, because there are many more companies (and capital) in the private space, especially as companies stay private for longer when the bar to IPO is so much higher today than two decade ago (e.g., $400MM ARR and $3bn-5bn valuation; The Real Data on What it Takes to Go Big and Eventually IPO with Meritech Capital | SaaStr).

Thanks for the content, Kara.

I've always been skeptical about applying this quantitative portfolio management techniques to PE investments. After reading the research paper, I also disagree with the authors on several assumptions, particularly the notion that the historical time-series volatility of a PE investment is equivalent to the cross-sectional volatility of available PE investments. This assumption fails to address issues like regime dependency and volatility laundering/smoothing that researchers studying this phenomenon such as Marcato/Key or Geltner desmoothing methods. 

In my view, the most valuable application of data science and ML/AI in PE is likely in analyzing alternative data to better understand the underlying state of companies and their key business drivers, which probably will allow PE firms to unlock value in early-stage funding rounds as it's  been happening over the last 20 years where IPOs have become less relevant to PE investors to exit their investments.

Great content, nonetheless!

------------------------------
Carlos Salas
Portfolio Manager & Freelance Investment Research Consultant
------------------------------

Really interesting articles. What is not covered are the perspectives of VC fund investors. Often investors are allocating to dozens of different funds and I've even reviewed portfolios with 100+ commitments. That is an enormous amount of underlying portfolio companies and cash flows (broken down intro many segregated different layers of fees, distributions and calls). Analysing the broader PE & VC market using AI to better plan your commitments, cash flows and allocations to VC is something institutional investors could do a much better job with.