Investment Strategies / Mechanical Investing❤
No. of Recommendations: 27
As a follow up to the discussion "at what point are you pretty darned sure your MI is working", I thought I'd start a thread where I'd post the results of my recent MI reboot.
My intent is to just keep adding to this thread, if I remember, rather than starting a new thread for each periodic snapshot.
A quick recap: I'd been out of traditional MI for a while, so I set up a new account and will track results using the total portfolio balance.
The portfolio is heavy on large caps and (oddly for me) relatively heavy on dividend payers. I pay 30% tax on US source dividends withheld at source, so that is baked into the result of the portfolio balance. Because of the emphasis on large caps, I'll use the S&P 500 equal weight index as the benchmark, tracked by RSP total return *without* a dividend tax. My bogey is "a monkey with a dartboard filled mostly with US large caps".
Since it's a real world portfolio, it's never quite 100% allocated, so there will also be a small cash drag. Of course all trading costs accounted for in the account balance.
It's a mix of screens, about 3/4 very conservative "about the same as the market but hoping for a bit better", and about 1/4 more zoomy/growth/momentum stuff for leavening. All stocks equally weighted, at least at the moment. Duplicate picks are only bought once, same weight as everything else. All stocks chosen from the VL 1700 universe.
Anyway, this is the result from the first three months.
Average about 4.0% cash to date. The portfolio has had cash taxes deducted equal to 0.10% of initial portfolio value.
Portfolio: Up 3.23% (13.09%/year annualized linearly)
S&P equal weight: Up 1.13% (4.44%/year annualized linearly)
Advantage relative to market: +2.20% (+8.65%/year annualized linearly)
If I do this for a few years, at some point maybe someone will figure it's time to chime in and say "OK, it's now pretty definite to me that your portfolio [is/is not] really adding value".
One quirky metric I will keep my eye on: divide the history into N equal-length time periods--what is the largest value of N for which the portfolio beat the benchmark in N-1 of the sub periods? Or all of the N sub periods. Ideally this exercise would require a history of daily balances which I don't really have a simple way to maintain, but I'll have monthly.
Jim
No. of Recommendations: 4
Just in time for the bear market?
This bull market is getting really long in the tooth. I worry about it every day.* Especially when everybody mentions that all the major indexes are dominated by the same dozen or so companies.
FWIW, I just got another alert from Fidelity: "52 Week High Alert: SPY"
---------------------
* "Bull markets climb a wall of worry."
No. of Recommendations: 8
Just in time for the bear market?
Yes, that's the idea!
In theory this portfolio is intended to a market tracker plus a bit, able to survive through thick and thin. (ha!)
I don't have a current plan to use a timing system (mechanical or tea leaves) to go to cash when things head down. My mood may change.
FWIW, I just got another alert from Fidelity: "52 Week High Alert: SPY"
It's a fresh all time high, not just 52 weeks : )
Either way, the "99 day rule" will be bullish for a while yet. A pretty good omen, if you like omens.
Jim
No. of Recommendations: 4
Just in time for the bear market?
Yes, that's the idea!
My 2 cents - The market hitting all time highs is NOT a time to bet on a bear market. With the European central banks cutting interest rates there is pressure on the fed to do the same, which is likely a positive omen for the market.
As for using an equal weight S&P index as your benchmark, that gives you an artificial advantage at present. The mega caps are significantly out running the equal weight index and have been doing so for some time, and it sounds like you're over weighted in the mega caps.
Elan
No. of Recommendations: 6
As for using an equal weight S&P index as your benchmark, that gives you an artificial advantage at present. The mega caps are significantly out running the equal weight index and have been doing so for some time, and it sounds like you're over weighted in the mega caps.
Mmm, maybe.
The stock prices of the very largest firms are doing very well lately, it's true, but on the other hand most of them are pretty profitable as well. There is apparent overvaluation ("prepaying for a bright future") in places for sure, but it's not particularly wild relative to the rest of the market.
In any case I'm not concentrated in super-giga-caps, the top 10 where the exuberant action (and cap weight index) has been. My holding's median market cap is about $40bn, and only 6% of the portfolio is in stocks over $300bn. Neither figure is much different from the S&P 500 list of stocks (and wildly smaller than the cap-weight index). That's why I figure the S&P equal weight is a reasonable benchmark.
Jim
No. of Recommendations: 0
Excellent. Seeing this, I got curious about IB's reporting which I have not looked at much.
I had been at FolioFN but IB scooped them up.
IB provides a very nice account summary through their "activity" report for any date range.
It shows overall return as well as commissions, interest, withdrawals, and individual holdings.
Most of my accounts have a mix of long term holdings and MI screens, and it would be more work to separate them out.
But you can download a CSV which has the information needed to do that.
Mark
No. of Recommendations: 2
In any case I'm not concentrated in super-giga-caps, the top 10 where the exuberant action (and cap weight index) has been.
Oh, okay. I thought you might be using your ROE_Cash screen which is heavily weighted in the giga-caps (or should we start calling them the tera-caps?).
Elan
No. of Recommendations: 3
Oh, okay. I thought you might be using your ROE_Cash screen which is heavily weighted in the giga-caps (or should we start calling them the tera-caps?).I can't remember, I think ROE_Cash is the same as the one I call LargeCapCash? Yup, I'm using that, but not
only that, for the conservative 75%.
I'm using a version of the "dividend required" alternative, which brings down the median market cap quite a bit.
FWIW, it has four years out of sample now. Original post
http://www.datahelper.com/mi/search.phtml?nofool=y...The original recommended use, 2 month trading cycle top 40 HTD 45, has returned after friction CAGR 26.3% versus 18.2% for S&P so far post discovery.
Unusually good, I wouldn't expect that much of a relative or absolute advantage over time. It has beaten the market for me in the short time I've been running it with real money, which is nice.
Jim
No. of Recommendations: 3
One quirky metric I will keep my eye on: divide the history into N equal-length time periods--what is the largest value of N for which the portfolio beat the benchmark in N-1 of the sub periods? Or all of the N sub periods. Ideally this exercise would require a history of daily balances which I don't really have a simple way to maintain, but I'll have monthly.
Oh wow! :-)
In another (not stock-related) area, I have a spreadsheet called nineheads. The idea being that instead of doing elaborate statistical calculations (I've never been much good at stats) I simply divide a data series into 9 pieces and calculate the up/down value for each piece. If I get 8 or 9 "up" I assume the series is reliably "up", and 0 or 1 (i.e. 8 or 9 "down") means reliably "down". The 6 levels between those extremes are various levels of "not sure".
I haven't really used it in anger yet, or offered it for criticism (don't think it would be well received) but I like the simplicity of it. I derived the "9" value (with the help of some internet stats tutorials and trial-and-error spreadsheets). Seems like you're doing a similar but more generic thing with a variable N, so your "score" is the size of N, and my "score" is the number out of 9.
That's made my day, no, my year!!! :D
SA
No. of Recommendations: 8
Seems like you're doing a similar but more generic thing with a variable N, so your "score" is the size of N, and my "score" is the number out of 9.
Yup, same idea.
Somehow I think a strategy that beat its benchmark for (say) 7 out of 8 years or 8/8 is better than one with the same CAGR but got all its advantage in the last year.
One of my screens is designed to be good that way. It specifically targets tracking the market fairly closely, just aiming for a small and relatively constant edge. The sort of result you would get by buying the entire universe of stocks, except simply skipping a whole bunch that are likely duds.
In fact I have a screen like that too, though I'm not using it: it's basically a long list of crap filters.
Jim
