3 ms·
> "Compare that to all the other 20 year windows. So unless you think this 20 year period is useful for predicting the next 20 year window, it’s not very helpfu
by autokad 6y ago
> "Compare that to all the other 20 year windows. So unless you think this 20 year period is useful for predicting the next 20 year window, it’s not very helpful."
No window is good at predicting any asset. As in, previous returns can't be used to predict anything.
Nobody cares about 50 year windows if that is well outside of their investment window. realistically, a person has about 20 years to build up assets before they need to switch to more safer bonds so talking about 50 year windows is like me saying I want to live in this geography for nice weather and you saying 'well over geological epochs, this place is far more stable'.
I think the sp500 should be most people's primary investment. but I also hate people who say the sp500 is infallible, when they used research that only looked at 1952 - 1999. so when you see people parrot 'the sp500 beats all asset classes over any long period of time'. well 20 years is a long period of time.
- prepend 6y agoI’ve never met anyone who says s&p is infallible, it’s usually that it is better than any alternative. I want a 50 year horizon for intergenerational wealth. For myself, I started at 20 and probably wont retire until 65-75. Although it won’t all be in s&p at 75, some of it will. The 20 year window is useful for retirement, but the fact that one 20 year window in the past 100 years out performed isn’t that handy as the s&p usually outperforms gold.