No. of Recommendations: 4
http://www.philosophicaleconomics.com/2016/01/gtt/

creates a timing rule tied to retail sales and production downturns as superior to MA200 rule
Print the post Back To Top
No. of Recommendations: 0
http://www.philosophicaleconomics.com/2016/01/gtt/

creates a timing rule tied to retail sales and production downturns as superior to MA200 rule


musselmant, this piece seems well done and addresses the questions in some of the other threads directly: Namely that a fundamental overlay to MA200 is a good idea.

Interestingly, the author determines that a CAPE overlay is not a good choice, though I'm not clear he used the correct threshold.

The point appears also well made that the risk adjusted return and MDD are also improved with this overlay.

Anyone see any issues in this author's logic?

(It's a long piece, but the meat is in this chart: http://i0.wp.com/www.philosophicaleconomics.com/wp-content/u... which shows the combined overlay approach "GTT" outperforming the MA200 approach alone "MMA" for total return as well as risk)
Print the post Back To Top
No. of Recommendations: 0
The point appears also well made that the risk adjusted return and MDD are also improved with this overlay.

Anyone see any issues in this author's logic?


TLDR

Beware of strategies that improve returns while lowering risks too.

Bryan
Print the post Back To Top
No. of Recommendations: 1
He uses FRED data, which is adjusted often later, so his test is probably not a real-time test as investors would do in real time using data as available at the time. How much this changes the results is thus unknown. But I take the general point that most stock downturns are related to recessions so the general point may be good, since the stock market is a leading indicator. I wish he had gone back and looked at FRED as it existed in real time.
Print the post Back To Top
No. of Recommendations: 0
TLDR

Beware of strategies that improve returns while lowering risks too.

Bryan


Why? Are you suggesting that it's not possible to do so?

-LB
Print the post Back To Top
No. of Recommendations: 0
Why? Are you suggesting that it's not possible to do so?

It is quite possible. Especially in backtests.

Timing shows the same possibilities.

Real world real time different story.

Bryan
Print the post Back To Top
No. of Recommendations: 0
Beware of strategies that improve returns while lowering risks too.

You still don't get it.

It's not about lowering risk, it's about lowering volatility.

Volatility is not risk. Took me a long time to get that through my thick head, but until I did, I kept making the same class of mistakes in understanding things.


We all have our blind spots, and it's useful to recognize them. As a wise man once said, "A man's got to know his limitations."

Don't take the following quotes the wrong way, I'm not trying to insult you, just trying to make the point to you.

"How do you explain to people who just don't get it that the problem is they just don't get it?"

"I can _explain_ it to you, but I can't _understand_ it for you."
Print the post Back To Top
No. of Recommendations: 0
It is quite possible. Especially in backtests.

Timing shows the same possibilities.

Real world real time different story.


Link with historical backtested data and then data moving forward. Short story, S&P returns with 1/2 the volatility.

http://mebfaber.com/2013/06/04/qtaa-paper-update-introductio...

JLC
Print the post Back To Top