Earlybird vs. Any in Startup Investment Models

Table of Links

Abstract and 1. Introduction

2 Related works

3 Dataset Overview, Preprocessing, and Features

3.1 Successful Companies Dataset and 3.2 Unsuccessful Companies Dataset

3.3 Features

4 Model Training, Evaluation, and Portfolio Simulation and 4.1 Backtest

4.2 Backtest settings

4.3 Results

4.4 Capital Growth

5 Other approaches

5.1 Investors ranking model

5.2 Founders ranking model and 5.3 Unicorn recommendation model

6 Conclusion

7 Further Research, References and Appendix

4.2 Backtest settings

In this study, several experiments were conducted with different backtest configurations, we called them earlybird and any. The earlybird configuration exclusively permits entry for companies only in rounds B or C, while the any configuration broadens the entry criteria to any round within the list of series_b,series_c,series_d,series_e,series_f, series_g,series_h,series_j,series_i, as long as they are within the considered backtest window.

The choice of entry configuration depends on the stage at which we enter the company. Similarly, the choice of exit configuration depends on when we decide to exit the company based on its success event (IPO/ACQ/unicorn), as discussed in Section 2.2. However, since the "unicorn" status can occur in the early rounds, there is a question of which round to exit. Two approaches were considered: using first approach we exit the company when the first success event occurs, while using last approach we exit on the last success event, analogous to "we sit until the end."

The main approach used in this study is the earlybird_last due to business requirements. However, this approach has its drawbacks, such as the fact that the company success flag becomes known later in time, resulting in a smaller dataset size for training at the beginning of the backtest and a slightly lower quality of the backtest compared to the earlybird_first approach.