The Lindy effect, a statistical phenomenon, suggests that things with longer histories tend to have longer futures. While this effect has been observed in some categories, it does not apply universally, leaving room for inquiries about its applicability and reasons behind it. In this paper, I explore the mathematical properties that underlie the Lindy effect and propose mechanisms that can generate them. Contrary to popular belief, the Lindy effect can occur even in cases with a constant or increasing hazard rate, as long as there is a probability distribution over the size of that rate. This implies that even entities becoming less robust with time can still exhibit the Lindy effect.
https://arxiv.org/abs/2308.09045