If you do something similar on bonds it will look the same. The next decade, perhaps 2 decades, are going to be awful for asset returns. This will have profound implications for the industryhttps://twitter.com/SoberLook/status/1460190968683147273 …
-
What happens when you use years ending in 1?
For the CAPE plot? I don’t know, I don’t have the data easily to hand. AQR redid the analysis a while ago and concluded that it’s not statistically significant if you do the analysis properly. My own view is that it probably has some weak predictive power.
-
-
Shpuld be just spx returns from xxx1 to yyy1 and the shiller pe in year xxx1, right?
-
Yep exactly.
End of conversation
New conversation -
-
-
The benefit of overlapping data is so minimal that even with 100 years and looking 10y returns, the effective number of samples compared to non-overlapping only increases by 2 years, as in 12 observations. Here is a Link and pic from the AQR paper https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3142575 …pic.twitter.com/V3Jtf5f8kk
-
The whole swarming effect, mostly due the high autocorrelation caused by overlapping data paints a very misconstrued picture.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.