So what are your results? BTW, when I clicked the link, Google gave me this warning (just the text, I'm too lazy to do a screenshot and upload it): Reported Attack Page! This web page at www.theivyportfolio.com has been reported as an attack page and has been blocked based on your security preferences. Attack pages try to install programs that steal private information, use your computer to attack others, or damage your system. Some attack pages intentionally distribute harmful software, but many are compromised without the knowledge or permission of their owners.
Re. the Ivy Portfolio, I've never had a problem with the website, but here's another with the same info, the author's blog: http://www.mebanefaber.com/timing-model/ The strategy was public a couple years before the book was published, I followed it in part in my 401K since early 2008, saved me some money then. An earlier comment indicated that they were checking daily. Generally that hurts results due to more whipsaws. This site has some nice little tools that let you test this easily for yourself: http://www.etfreplay.com/backtest_ma.aspx
Thanks for the links! I'm guessing in most cases, trading on daily signals with the 200 day is going to create many more false signals that will hurt performance. Perhaps this sort of logic could be applied to moving averages in different time frames to reduce whipsaws and improve performance. I had seen an article regarding this type of allocation/timing strategy in SFO magazine, and done some checking. It seemed much superior to trading MA's on every signal. I guess I got lazy and didn't do any follow-up however.
Last year, 18%. This year, so far 13% annualized. My backtesting indicates that 15% is going to be about average. As you might expect, about half of trades are losers. Winners are much larger than losers. Average holding period is about 3 months. I backtested several MAs and decided that if I wanted to use only one rule, the 200 and the 50 constituted the best compromise. Maybe there is a reason why these are the "default" MAs. They seem to work best in most situations. And I think it is best to pick a set of MAs and stick to them. Otherwise there is too much curve-fitting. I take most, but not all, signals. For example, if I am long the Aussie and the Loonie posts a signal, I do not buy it, so that I do not hold two commodity currencies. Therefore, when there are highly correlated signals of this type, I end up taking the one that has the first signal. My impression is that this works best on commodity, bond and currency ETFs. Stock ETFs, not so well--they are too volatile and too bandied about by earnings and news reports. As with any trend following method, there are some periods with a lot of whipsaws. You gotta stick with your rules.
A little bit of whipsawing isn't bad for ETF trading, like $8.95 at Schwab, plus I've heard the some brokers are offering commission-free ETF trades. ETA: From the 9/14 entry on Faber's blog: http://www.mebanefaber.com/
instead of the junk bonds I would get into munis or better a muni index at vanguard. At any rate if you don't use vanguard you'd be burning money, at VG you are part owner and in fees you'd be saving a ton. This is not easy because to get a grand a month (SAFELY) you'd need more money. Like a few zero at the end
Must the return come on a monthly basis or annually is ok ? It's possible to achieve those results when volatility is high if you give time some flexibility, but you do need volatility. ESD