I am pretty happy now that I have monthly data going back to 1972 because I do a lot more testing of certain concepts.
The 15/35 rebalancing bands are interesting because they introduce a momentum effect. Versus monthly rebalancing, with the 15/35 bands you end up holding more of what did better in the recent past and less of what did the worst. Historically this momentum affect has been positive.
You can also see the cumulative out-performance by expressing it as a ratio.
It is interesting how much of the out-performance came from the late 90s.
Aside from the momentum effect, I like that the 15/35 bands involve less trading. It allows you to reduce your costs and if the momentum effect continues to work in the future that is just a nice bonus in my mind
Last edited by melveyr on Sat Mar 23, 2013 5:44 pm, edited 1 time in total.
everything comes from somewhere and everything goes somewhere
We've had a few threads debating between 20/30 and 15/35 bands. A few of us (me included) ran some tests and found that 15/35 performed better. I thought the momentum effect was responsible for that, but these charts prove it clearly. Thanks!!
"Democracy is two wolves and a lamb voting on what to have for lunch." -- Benjamin Franklin
With all this data, have you look at what is the best new contribution method? (I think there have been other discussions on this question, but shorter term.)
1) buy the lowest asset
2) equal contributions into each assest
3) cash till you hit a rebalancing band
I do #1 monthly, it make me feel like I am getting a deal, but I think #2 has won on shorter simulations.
“Let every man divide his money into three parts, and invest a third in land, a third in business and a third let him keep by him in reserve.� ~Talmud
melveyr wrote:
I am pretty happy now that I have monthly data going back to 1972 because I do a lot more testing of certain concepts.
You weren't happy with simba's spreadsheet data?
In my backtesting, I have found that quarterly and annual rebalancing seemed best. I favor annual due to tax considerations. Monthly and weekly just never seemed to add anything. Neither does rebalancing around the seasonal effect. I'm not able to compare band rebalancing and time rebalancing in the same software against each other without low-level programming. Perhaps you can do it.
Be sure you subtract some kind of realistic commission costs when doing these rebalancing comparisons. $75 each way for mutual funds ala current Fidelity or TDA is likely realistic for the past 40 years. Brokerage commissions used to be fixed at $120 minimum before May Day. TANSTAAFL.
Last edited by MachineGhost on Sun Mar 24, 2013 6:39 am, edited 1 time in total.
"All generous minds have a horror of what are commonly called 'Facts'. They are the brute beasts of the intellectual domain." -- Thomas Hobbes
Disclaimer: I am not a broker, dealer, investment advisor, physician, theologian or prophet. I should not be considered as legally permitted to render such advice!
Glad to see somebody else mention the word "momentum". I remember using it not long ago in a discussion about buying lagging assets only to find myself accused of being a market timer. (Is there anything worse that you could call a PP'er)?
I seem to have some sort of mental disability that prohibits me from understanding charts very well but it just seems intuitive to me that if something isn't broke then you shouldn't try to fix it. And I would define "broke" as hitting the re-balancing bands.
I find myself currently in the interesting position of experimenting with doing the complete opposite of buying lagging assets, although I hadn't planned it that way. All my new money is going into stocks by 401k necessity. It will be interesting to see how it turns out.
melveyr wrote:
I am pretty happy now that I have monthly data going back to 1972 because I do a lot more testing of certain concepts.
You weren't happy with simba's spreadsheet data?
I think looking at annual data can hide a lot of risks. The PP has had a couple of really brutal gold related drawdowns that don't reveal themselves when looking at annual data, but can be visible when looking at monthly data. Since we all experience the PP intra year (unless someone here is only looking once a year), I find it relevant to see what is historically possible.
everything comes from somewhere and everything goes somewhere
Consider the effects of investment costs with monthly rebalancing. If you can do this without paying fees, great. But buying/selling small increment investments can mean some hefty fees over time.
