[Paper Review] Probabilistic Modeling of Venture Capital Portfolio Outliers

Kensei Sakamoto, H. U. Köylüoglu+2 more|arXiv (Cornell University)|Feb 8, 2026

Private Equity and Venture Capital0 citations

TL;DR

This paper builds a latent-factor Gaussian model to quantify correlated venture outcomes and outlier risk, demonstrating that expected unicorn counts are insufficient and that correlation structurally affects portfolio reliability and upside.

ABSTRACT

In this paper, we define probabilistic measures for venture portfolio performance based on individual outlier probability for each investment and the dependence across investments. This work is inspired by loan portfolio modeling against default risk used in banking. In mathematical terms, we calculate the probability distribution of the sum of N non-homogeneous Boolean outcomes (investments becoming outliers) that are correlated through common factors such as overall market conditions and sector effects. Specifically, we implemented a latent-factor model in which each investment's success is the exceedance of a Gaussian latent variable composed of idiosyncratic returns and returns from interpretable shared factors (stock markets, industry sector indices, geography and founder type). The formulation follows a simulation approach to preserve heterogeneous deal-level success probabilities and uses empirically estimated correlation matrices. When applied to synthetic portfolios, our model reveals that expected outlier counts alone are insufficient statistics for evaluating venture portfolios. Portfolios with identical expected outcomes can exhibit drastically different levels of reliability and risk when various levels and forms of correlation are embedded. Diversification improves the probability of achieving a minimum number of outliers by reducing exposure to common shocks, but at the cost of lower upside, underscoring a fundamental tradeoff between reliability and magnitude of clustered successes. The framework provides a practical bridge between deal-level outlier probability assessment and objective-aware portfolio construction.

Motivation & Objective

• Motivate VC portfolio assessment beyond mean returns by focusing on rare outlier successes (unicorns).
• Introduce a flexible latent-variable framework with sector, geography, and founder-type exposures to capture dependence among investments.
• Preserve heterogeneous deal-level success probabilities while embedding interpretable cross-investment correlations.
• Demonstrate, via simulations, how correlation alters left-tail risk and optimal diversification strategies.

Proposed method

• Define A_i as a Gaussian-exceedance latent variable with A_i = w_i^T Z + sqrt(1 - w_i^T Σ w_i) ε_i.
• Model unicorns as events A_i > Φ^{-1}(1 - p_i), preserving individual success probabilities p_i.
• Encode sector, geography, founder-type affiliations into w_i via loading construction with weights (S, G, F) = (0.6, 0.3, 0.1) and normalize to b_i with w_i = w_0 b_i.
• Set w_0 to target average pairwise correlation ~0.12 using ρ_{ij} = w_i^T Σ w_j and Σ estimated from public proxies.
• Estimate Σ from monthly returns (2020–2025) across 5 sectors, 4 geographies, 2 founder types, using Cholesky sampling for Z ~ N(0,Σ).
• Use Monte Carlo simulations to generate correlated unicorn outcomes for portfolios and analyze distributional properties.

Experimental results

Research questions

• RQ1 How does correlation across VC investments affect the distribution of unicorn counts beyond the mean?
• RQ2 What is the impact of diversification on left-tail risk (P(U=0), P(U≤1), P(U≤2)) under correlated outcomes?
• RQ3 How do portfolio composition (sector/geography/founder mix) and size interact with correlation to shape risk and upside?
• RQ4 What are the tradeoffs between reliability (left-tail risk) and magnitude of clustered outlier successes (upsides) under dependence?

Key findings

• Correlation increases the probability of joint failures and successes compared to independence for a fixed per-deal probability.
• Portfolio diversification reduces left-tail risk but can lower upside once thresholds are reached.
• Full multi-factor correlation modestly worsens left-tail risk relative to single-factor, even with sector diversification.
• Increasing standalone success probabilities under correlation yields slower reductions in left-tail risk than under independence.
• Portfolios with identical expected unicorn counts can have very different reliability due to dependence structure.
• Optimal portfolio design under dependence depends on whether the objective is minimizing failure probability or maximizing conditional upside.