Hacker News new | past | comments | ask | show | jobs | submit login

If you were able to buy those numbers for the last hundred years, you’d have <$25k invested to make $200M.

It’s no more absurd than projecting stock returns over 100 years.




The difference is the stock market in not a game with net negative expected value


That’s impossible to say.


I see two options (or, 3 options, depending on how you count it?): Either we interpret this taking probability to refer to some objective probability distribution, or one takes it to refer to a subjective probability distribution.

In the first case, uh, one can either talk about the empirical distribution, and like, because the downside is limited ("all the money one invested in it"), if one puts p-value-ish confidence intervals on it (I don't know the right way to talk about this) on it, and, assuming we treat the behavior at different times in history as comparable, or like, if we assume that the current moment is randomly sampled from the time in which the stock market exists, or something like that, uh, I imagine that it would be possible to say something like "If the stock market across time-as-a-whole and like, selecting when you buy in and cash out at random, with like, some bound on how far apart those two are, had a negative expectation value, then we would see behavior like this with probability less than p" ? I don't know what the value of p would be, but, I suspect it would be fairly small, at least, for some ways of formulating the statement.

Or, one could take a subjective probability distribution view of, uh, what the unknown objective distribution is, and so the statement that it has a positive expected value is just a statement that one assigns a high probability to it having a positive expected value.

Or, one could just take a subjective probability distribution view of like, how it will behave, and interpret the statement as subjectively assigning positive expected value of investing in the stock market.

I think? This seems to make sense to me, but, I've not like, read much about philosophy of probability or whatever, and also I could be missing (or wrong about) some of the math.

But, in any of these 2 (or 3) cases, it doesn't make sense to me to say that it is "impossible to say" whether "the stock market in not a game with net negative expected value".




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: