Forecasting winner IPOs

ABSTRACT

IPOs underperform on average. Nevertheless, the skewed distribution of returns offers the chance to gain extremely high rewards (e.g., identifying the “next Microsoft“, as discussed by Loughran and Ritter, 1995). Hinging on this argument, this paper proposes a new method to help investors screen IPOs for the high-performing tail of the returns distribution. Using a straightforward definition of “winner IPOs” based on buy-and-hold abnormal returns, this study employs logistic regression to forecast whether a firm is still a top performer 1, 2 or 3 years after listing, relying only on publicly available information. Investors using our forecasting model would always have an adjusted rate of successful predictions higher compared to a naïvely classification that consider all IPOs as “winners”.