"It's been beat into the ground already, but the short version is that it pretty much doesn't work. The few studies that have looked at it have looked at the effect in very narrow scenarios within the domain of real estate, but they find a large effect, and no reason to expect it to fade out.
"The long version is that anyone investing in domains today who thinks that buying a domain in one of these one-off studies is likely to get very high returns and do very well in the long term is almost certainly going to be taken to the cleaners. There are many domains that priced very low because nobody who knows anything about the space wants to deal with them and they have no monetization path because they cover niche content that nobody cares about. So buying domains like that is unlikely to work out. And there are other domains that buyers think are cheap because some studier showed they had a high return, but that's because neither they nor anyone else knows that these domains have no monetization path. So buying those is also unlikely to work out in the long term."