Beyond Gut Feel: Using Time Series Transformers to Find Investment Gems

TL;DR

This paper addresses the growing application of data-driven approaches within the Private Equity industry, particularly in sourcing investment targets, and proposes a novel approach leveraging a Transformer-based Multivariate Time Series Classifier (TMTSC) for predicting the success likelihood of any candidate company.

Abstract

This paper addresses the growing application of data-driven approaches within the Private Equity (PE) industry, particularly in sourcing investment targets (i.e., companies) for Venture Capital (VC) and Growth Capital (GC). We present a comprehensive review of the relevant approaches and propose a novel approach leveraging a Transformer-based Multivariate Time Series Classifier (TMTSC) for predicting the success likelihood of any candidate company. The objective of our research is to optimize sourcing performance for VC and GC investments by formally defining the sourcing problem as a multivariate time series classification task. We consecutively introduce the key components of our implementation which collectively contribute to the successful application of TMTSC in VC/GC sourcing: input features, model architecture, optimization target, and investor-centric data processing. Our extensive experiments on two real-world investment tasks, benchmarked towards three popular baselines, demonstrate the effectiveness of our approach in improving decision making within the VC and GC industry.

Figures (7)

• Figure 1: An illustration of multivariate time series dataset ($N$ samples) to train models for VC and GC sourcing.
• Figure 2: TMTSC architecture: $\mathbf{u}_t$ and $\mathbf{v}_t$ are numerical and categorical part respectively.
• Figure 3: U-GRU: each univariate time series is modeled by a BiGRU block.
• Figure 4: M-GRU: all time series features are modeled by one single BiGRU block.
• Figure 5: TE architecture with 4 Transformer encoder blocks.