Ok, it is about time I got around to adding the FAQs to the website. If you have a burning question, or anything you want answered regarding the timing paper, the book, AlphaClone, or anything else place it in the comments and I’ll include it in the FAQs.
Have a great 4th!
I know there is no majic bullet answer to this one, but…best/better ways to approach entries at 10 month cross of price–scale in, all at once, wait until end of monthly close above MA, high momentum cross, first daily pull back, etc.
Thanks.
Sometimes the description of the methods are not entirely clear. A recent example is the “hedged rotation”. I had always thought that hedged meant “…technique of taking two positions that will offset each other if prices change”. My understanding was, of course, incomplete.
It took several readings of the post before I figured out what you meant.
In the FAQ, it would be nice to detail the rules (for each variation) laid out in steps 1,2,3, etc.
Re alphaclone,
I am a subscriber. In calculating the various returns, why not use the dates when the holdings become available to subscribers, ie, 45 days after quarter?
Thanks.
Another question. In a 401k, since shorting is not allowed and using short etfs eats up half the assets, what is best way to hedge? Puts? And if so, long dated; at the money; out of the money?
Or where might I read up on ideas.
Thanks.
Q: Why is Meb such a badass?
The timing model mentions 2x leveraged model. What would be some simple low cost methods for an individual investor to implement this? Obviously the leveraged etfs won't do the trick, since they are 2x *daily* returns.
and another couple of Qs from me
- What are the best criticisms you've read or come across of the timing model, or any of your models for that matter?
- In what environments are your models expected to underperform, and more importantly – do you have ideas for identifying such environments as they happen (so kind of “timing the model” if you will).
Please disregard this question!!
Mebane, any plans on doing a junk bond backtest?
Answer: Self-evident.
Any suggestions of Euro ETFs in order to invest in the “Timing Sistem” for european individual investors?
thx
Meb – For the FAQs: Do you consider the effect of distributions in your 10-month calculations or simply the absolute price?
do not understand what “hedged rotation” means. if not going to cash, you mean to buy inverse ETFs? and when should you start to hedge and close it?
It seems that in some of your literature I read that waiting 2 months or until the 200dma crosses the 240dma ensured greater stability of the uptrend. Did I understand this correctly?
For the rotation system you've written about where you purchase the top performer over the past 3, 6, 12 month periods, are you simply using the mean of the 3, 6, 12 month performance to calculate the top performer?
Mebane: I greatly appreciate your work and everything you have provided for us.
Question #1 – In the timing paper, your buy/sell signals are based entirely on the security's piercing of the 10-Month Simple Moving Average. Would an Exponential Moving Average change and/or improve the results? Would using another moving average with a shorter time frame – 2-Month or 3-Month – to be applied in a crossover manner (like the popular 50-day crossover of the 200-day forming a “Golden Cross”) for creating buy/sell signals, improve results?
Question #2 – Will you be publishing Part 2 of Theodore Wongs research (Moving Average: Holy Grail or Fairy Tale)? Did he find that 10-Months is, in fact, the optimal time frame for the MAC system?
Thank you.
regarding holy grail or fairy tail:
part 2 can be found on the guys website
great read!
and his conclusions are still valid
but the limitations as I expected will be discovered in part 3 where he will post the decale performances. I think that in the stable bear markets of the sixties the 1M 10M MA will not prove to be as good due to several false signals in one decade…
Can't wait till it comes out
Why doesn't glue stick to the inside of the bottle?
Where do forest rangers go to “get away from it all”?
Why isn't there mouse flavored cat food?
Can this system works if you don't use it in a tax advantaged account?
Does the system still work if you buy or sell one the signal is given or at the end of the month?
what would the CAGR, Largest Drawdown and Std Deviation be like if we invest Top 3 (if all 3 is above the 10mth m.a.) and switch to Top 2 (if just 2 is above the 10mth m.a.) and just Top 1 if only one asset is above the 10mth m.a. and be totally in cash (3 mth T-bills if none are above the 10mth m.a.)…If this has been answered somewhere here, can someone please post the link. thanks for sharing your research.
Is the 10 month sma crossover optimized for all (five) asset classes, or is is possible that different timeframes could work better for different asset classes?
Thanks, Mebane
Sorry, one more, Meb. What is your opinion on using emas instead of smas? Does one improve the timing results over the other?
Thanks.
has anyone here ever red 'smarter investing in any economy'. I don't get his ranking formula forrelative strength on page 42 where he divides the percentage change of an asset by the percentage change of the peer group (index). I mean if the asset is down 50% and the peer is down 25 %then you get 2 as a result…which is a positive value although the asset underperformed the peer…totally wrong…
even worse:
what if the peer change is 0 %.You can't divide by 0
I really don't get it how he got this one so wrong…
How does one use a mutual fund (or ETF) in the timing system: when a dividend is paid in a mutual fund it can greatly affect the price, pushing it to the 200 DMA, when in fact the fund is doing well. Is there a simple way to account for this?
This post probably repeats points made by others.
1. In a rotation system, should one simply average the results from 3, 6 and 12 months returns of the 5 ETF catergories (or 10 or 20 if using the more granular categories) and the highest is # 1, second highest # 2, etc.; or are they weighted in any way; or all made equivalent to a yearly return and then averaged; etc.? The exact calculation would be helpful via examples.
2. If one uses a more granular approach, it would seem that 3 categories might not be best since they could all come from within one of the 5 original groupings. This would not be very diversified.
3. I downloaded ETF info from FundX's website (it's an investment newsletter) and compared the 3, 6 month calculation with their “CPR” calculation. They are highly correlated and led to practically identical results. One could use their CPR rating to determine which ETFs are the rotation leaders. Since they rate many ETFs (though not all), this would save a lot of work.
4. Could you use an ETF such as FXI (China) as a portion of a category rather than the generalized index ETF? I would limit a narrow ETF such as FXI to 20% of the generalized category for reasons of diversification. I would always use the index if the 3, 6 and 12 month returns average were higher however.
5. IT would be good to list the best websites to gather all the necessary info for the rotation system. As I mentioned, FundX is good but it costs money. Not every ETF is listed at FundX. VB is missing for example.
5. Wong's paper indicated a preference for a 6 month (100 day) moving average and he preferred the EMA 100 over the SMA 100. What do you think?
6. What's the minimum daily averager trading that you think is adequate to keep the ETF liquid? I've seen $1,000,000/day or 40,000 shares mentioned.
7. May I suggest trailing stop losses at about 2.5 to 3.5 times the Average True Range? This would equal about 5 days of losses in a row. IT would sem to be sensible to use a multiple of the ATR – the question is what is the multiple?
8. After I bought the book and had time to absorb it and figure out the methodology I wondered about how to install the system for the first time. Buying one of the 5 ETFs if it were above the SMA 200 sounds simple but what if it were above the SMA 200 for a number of months already and is about to run its course? When entering the system should one buy when an ETF first goes above the SMA 200 or within a month or two of it going above it, or does it not matter? I'd guess it DOES matter. I find it hard to believe that buying an ETF that has been above its SMA200 for ten months is the same as buying one that just went above it.
9. This is a question involving a rotation system with a more granular approach: right now the VTI or S&P 500 is below its SMA 200. However, the small cap ETF might be above its SMA 200. Should one buy the small cap ETF?
The same question goes out for all the other categories: if DBC or DBP is below its SMA 200, but say GLD is above it, could one buy GLD?
10. I wonder if corporate bonds and government bonds should be separated into categories. I suspect that they will behave much differently in the future.
11. Is there a place for TBT (short long Gov. bonds) in this system?
Finally I want to thank Mebane for a good book and a practical guide to improving one's chances in the market.
Mebane,
Your work has been extremely helpful, especially the QATAA paper and the Ivy book. I manage my own stock portfolio but have been trying to lead our 28 doctor group’s deferred comp (retirement plan) out of this 08 bear mess they have gotten us into. It has been invested in an S&P index mutual fund. Trend following/technical analysis investing is nothing new to me and have been trying to teach it to the group. Early last summer I tried to get the board to go 100% to cash and was met with extreme skepticism and buy and hold dogma. Well by the end of the year they started listening to me. In trying to help derive an investment model for the fund I did extensive searches on the net and stumbled over your work and have been sharing it with our board. I pre-ordered the book and obtained the first copy out on Amazon and have shared the information with the board members. It helps to having someone gone through the data and run the numbers so that I don’t sound like some crazy stock trader in getting them to understand the markets and that “Buy and Hold is Dead” and needs to be buried!
We have a number of self-directed mutual funds available to us in our COLI (company owned life insurance) that we can utilize. We will be having a board meeting in a few weeks to vote on using the trend following and allocation model. Currently I have recommended 20% each in (all mutual funds) #1 International fund that tacks MSCI EAFE. #2 Emerging market fund. #3 A commodity mining fund – tracks the XME type. #4 S&P index fund. #5 Real Estate fund.
We will use the corresponding ETF/index for charting and use the 10-month moving average for the timing model. Since we can’t get a ticker on these we will us the corresponding benchmark index or matching ETF for charting. Currently will not include bonds but will probable add in say 5 years when some of us start retiring out of the group. (followed in by new members). I’m am also looking at putting in a Tech fund into the mix since in the new world order will probably have quite a bit of outperformance in the next 5 or 10 years. However I don’t want to get to far from the KISS principle that the others can’t keep up with.
Appreciate your site and your info.
Sincerely
GRock
Meb – thanks very much for your book and all the info you provide!
I am contributing monthly to my Fidelity Roth IRA, so am more interested in using mutual funds than ETFs to implement a 4-asset ivy portfolio (no bonds). I am considering moving the IRA to Vanguard to use their index funds without Fidelity's high commissions.
Can you suggest an index fund that is appropriate for the commodities portion of the portfolio? I'm having a hard time coming up with anything other than precious metals.
Second, with monthly contributions, is it better to evenly divide the contribution among the 4 portions, or allocate disproportionately to the portion furthest behind the 25% allocation?
Thanks again!
Ben personally I see no reason to use mutual funds instead of ETF type funds. Mutual funds have much much higher management fees as oppose to ETF's and all it will do is hurt your return compounded going forward. If your are attempting to model after the IVY portfolio I think this is the best way to go. My own comments above about using mutual funds is only because our group has no access to ETF's for our COLI plan. If you are using Fidelity you could just switch to their brokerage account and use the the ETF's.
As far as dollar cost averaging new money into the IRA you certainly could do equal amounts among the funds, or could put more in the laggard fund as a way to “rebalance” the portfolio. I kind of think it is best to do it equally since a hot sector will likely to stay hot for awhile and just rebalance on an annual basis.
GRock
Ah, but Fidelity charges $19.95 per trade online (at their lowest tier). If I'm purchasing 3 or 4 ETFs per month, in addition to moving between cash and holdings based on the moving average, that will add up fast. Mutual fund trades are free at fidelity. I was considering moving to vanguard for the low-fee index funds, but they don't seem to have any good commodity funds.
So, I think I've come back to ETFs, and am now researching free/inexpensive ways to buy them. So far, Bank of America looks most promising: http://www.bankofamerica.com/investing/overview...
Ben, certainly 19.95 is high for an online broker and certainly are cheaper alternatives. I don't know what kind of money we are talking about but you need to look at the frequency of trading in Meb's work. Generally if the market will give us significant trends before it rolls over your not bouncing in and out that frequently and even with a smaller account sizes I think you would be better off in an ETF. Generally If you read the fine print in mutual's prospectus you will find fees as high as 1.5% of the fund. In some index funds it will be as low as .5% which compounded adds up. You can do the math with your own account numbers. Certainly if you get a low cost brokerage account it becomes a no brainer. Be sure and to be aware using the moving average system that sometimes the market will get mixed up on it's own trend and sometimes the moves will be buy high and sell low for small nicks in one's account before one has a big buy low and sell high. I have looked at various historical markets like the dow and Nikki going back decades using various trend following methods to back test and get a feel how they would work. I did this before discovering Meb's work. I would encourage you to go through the Nikki with say Yahoo charting program going back to the 80's and test the moving average model to get a feel for it. Not to say we will have that market for the next 20 years but the market run we have had since 82 to 2000 won't likely occur again for a very long time. Certainly the bank of America's low cost brokerage is worth investigating.
Hope that helps>
GRock
Meb,
I'm a new Alphaclone subscriber and I was wondering if there is any data on using options to hedge clones? I can't go short in an IRA, but I can buy puts or put-spreads. More capital can be used for the actual clone picks with ATM index-puts acting as a hedge if prices decline. Plus, loss of capital is limited should the market move higher. Wouldn't the leveraged nature/limited loss of the hedge lead to better returns then going short SPY?
I'd like to buy options at rebalance that expire at the next rebalance. I just don't have the backtested results. I'd imagine it's possible to estimate the cost of hedging by knowing the VIX, days till expiration and SPY price of past dates. I'd love to experiment with put spreads.
Meb,
I'm a new Alphaclone subscriber and I was wondering if there is any data on using options to hedge clones? I can't go short in an IRA, but I can buy puts or put-spreads. More capital can be used for the actual clone picks with ATM index-puts acting as a hedge if prices decline. Plus, loss of capital is limited should the market move higher. Wouldn't the leveraged nature/limited loss of the hedge lead to better returns then going short SPY?
I'd like to buy options at rebalance that expire at the next rebalance. I just don't have the backtested results. I'd imagine it's possible to estimate the cost of hedging by knowing the VIX, days till expiration and SPY price of past dates. I'd love to experiment with put spreads.