Up about a tenth of a percent. The new portfolio is interesting. The foundation — the relationship between returns, volatility and correlation — is pretty standard right now. Slightly conservative, but not much. However, the standard deviation of everything is similarly elevated and a bit more correlated than usually.
That’s why the portfolio can’t shake the call for cash. The cash in this case doesn’t come directly from trying to optimize long term returns, but instead is the results of a factor of safety for errors and for drawdown control as the overall portfolio volatility is likely to be high. I will discuss both in future posts. On May 15th, 2020, the strategy rebalanced to:
41% SPY , 24% TLT , 20% GLD , 15% Cash
I’ve been reading the related comments on the boglehead thread. You’ve created quite a lively discussion! One comment in particular caught my attention. To paraphrase, Chi Capitalist made a point that
the bond carry estimates are clearly estimates of the geometric distribution, as in the case you hold till maturity that is in fact the return you get. And there is variance in the mean time. So the geometric return is the yield and the expected arithmetic distribution is actually that carry plus 0.5*variance.
Have you considered this? I noticed in your spreadsheet example the expected bond return is current yield + constant of 0.007, but it’s not clear if this was the logic and the constant reflects that approach.
My bad, you covered this on the Feb 16th post. Makes perfect sense.
No problem, thanks for reading and glad you found it.