While I agree that "messiness of markets can't be quantified accurately," I do think there's a risk of being so conservative that you didn't live the lifestyle you could have and then have regrets when you can no longer do those "fun" activities, but you're still cognizant enough to have regrets. I think using a Variable Percentage Withdrawal (VPW) rather than a constant-$ strategy helps mitigate the "didn't spend enough" concern, but that approach is also back-loaded (you draw the biggest percentage of portfolio in later years, not earlier years).I've no idea if my assumptions are conservative enough and could account for the messiness of markets that can't be quantified accurately. I guess time will tell...
I think arbitrarily increasing the standard deviation while reducing the average return (thus flattening the distribution and exaggerating the tails) by a small amount is a reasonable step to guess what if the future distribution is worse than the past distribution. Increasing the worst returns and reducing the best returns, when you already flattened the distribution and extended the tails seems like double-counting really bad bad outcomes, but up to you (it's certainly conservative!).
I think if you run several different sims and they all indicate "I'm going to be just fine in retirement, with acceptable risk to me," then you don't need to go overboard with making your own model arbitrarily worse in a big way (in a small way is fine).
Have fun with your modeling efforts (I certainly do!
)!!Statistics: Posted by bonesly — Sat Feb 01, 2025 10:04 pm — Replies 47 — Views 3499