> If you invested $1 in AIG at the start of 1990 and received only intraday returns
This is the problem with all stock market related articles: they all hinge on if you had done this or that. They never tell you I actually performed these two strategies and here are the results.
It would be quite simple to perform such experiments but it seems that it is either never done or is perhaps kept confidential.
In this case I'd expect the costs of trading in and out of the position everyday to be prohibitive. Even if your broker doesn't charge you anything (or you are a prop trading desk with zero trading fees), just paying the spread every day would add up quickly over time. It might not be possible to perform such an experiment at all in the real world.
It's still interesting from an academic standpoint though, if only to show how long term pricing inefficiencies can persist over long time periods even in otherwise efficient markets.
This is the problem with all stock market related articles: they all hinge on if you had done this or that. They never tell you I actually performed these two strategies and here are the results.
It would be quite simple to perform such experiments but it seems that it is either never done or is perhaps kept confidential.