Page 1 of 1

Calling all Backtesting Mavens

Posted: Mon Jul 01, 2013 12:35 pm
by glennds
Sometimes I see statistics on the market that say things like: X is what your return would have been investing in SPY over a given period and X- is what would have been your return if you had missed the 10 best trading days during that same period. The main point being that the peaks and valleys heavily influence your return and that you should stay in for the long haul and not try to time the market.

Following this train of thought I have wondered what a Permanent Portfolio backtest would look like if the owner favored adding to the portfolio during periods of drawdown like we have now? I realize that orthodox PP theory is based on not timing the market and that the portfolio is designed to protect us from the dangers of doing so.

But could you have your cake and eat it too by taking the position that if the PP  is down then the collective group of assets is on sale, and what better time to add to your portfolio? I'm guessing that modeling a backtest based on this practice might amp up the return without much more volatility. Any thoughts?

Bottom line - I'm trying to decide for myself if now is an advantageous time to add to my PP.

Re: Calling all Backtesting Mavens

Posted: Mon Jul 01, 2013 1:05 pm
by Pointedstick
glennds wrote: Bottom line - I'm trying to decide for myself if now is an advantageous time to add to my PP.
Well it's certainly a better time to add now than it was 6 months ago IMHO. Back then, three assets were at all-time highs and now only one is (stocks).

Re: Calling all Backtesting Mavens

Posted: Mon Jul 01, 2013 1:21 pm
by MediumTex
Pointedstick wrote:
glennds wrote: Bottom line - I'm trying to decide for myself if now is an advantageous time to add to my PP.
Well it's certainly a better time to add now than it was 6 months ago IMHO. Back then, three assets were at all-time highs and now only one is (stocks).
An interesting PP strategy would be to calculate a rolling average of the number of days since the last all time high in the PP.  As the number of days increases, I would assume that it would be a better time to add more money to it since the PP is rarely more than a few months removed from an all-time high.

The formula might call for increasing contributions once the portfolio is 100 days removed from an all-time high, then another increase at 200 days, 300 days, etc.

This strategy is based on the assumption that the PP will always be moving to a new all-time high in the next 12-24 months.  This may not be a sound assumption going forward, but historically it always has been the case.

Re: Calling all Backtesting Mavens

Posted: Wed Jul 03, 2013 10:38 am
by blackomen
How about using the Turtle Trading Rules to move in/out of the PP?

http://bigpicture.typepad.com/comments/ ... erules.pdf

You could keep 50% of your portfolio invested in the PP all the time and the other 50% traded according to the system.

(Note: you don't trade gold, stocks, and bonds separately according to the Turtle rules although you could also try that..  you look at the PP as a whole and apply the rules.)