ETF Replay Portfolio Review: June 2013 [View article]
See ETFReplay's reply below, the rotation is the "stop". A stop may be more appropriate when trading individual companies, but when trading ETFs/asset classes I'm not a fan if you have a momentum system already in place.
The ETFs are ranked for each factor (Returns are ranked high to low. Volatility is considered a negative and is therefore ranked low to high). The factor ranks are then weighted to produce an overall rank. For example, assume we have 6 month returns weighted at 40%, 3 month at 30%, 3 month volatility at 30% and 15 ETFs. ETF "A" is ranked 5th out of 15 for 6 month returns, 3rd out of 15 for 3 month returns, and 10th out of 15 for 3 month volatility, its weighted rank would be .4 x 5 + .3 x 3 + .3 x 10
All Star Battle: Graham Vs. Piotroski [View article]
It was, and I could run some simulations going long/short. My impression is that the majority of individual investors / readers find shorting via index ETFs easier to implement
It's a simplified "risk parity" calculation intended for 100% allocation, and probably a matter of semantics. A more robust calculation would incorporate correlations and could use leverage to target volatility. For background on the risk parity calculation I used Empiritrage has a good example http://bit.ly/10oftYW and ETF Replay also uses the same simple calculation for the "risk parity" backtest function on their site.
Keep in mind a few things: the allocations change over time as correlations and volatility change. TLT is the most interest rate sensitive position and has a low weighting at present due to its high volatility. High yield bonds tend to correlate more with equities than Treasuries (HYG/TLT presently have a correlation < 0 while HYG/VTI have a correlation > .5) and EMB also has low historical correlation to TLT. The 8 securities presented may not be an optimal portfolio, but it does serve as a framework for showcasing alternative methods for asset allocation. AGG has a higher correlation to TLT and it is also a broad-based US bond fund, so investors worried about inflation could seek to diversify these two positions. However, I think a viable benchmark for the portfolios in the article are the commonly used, static 60/40 (stocks/bonds), 50/50, 40/60 asset allocation which maintains a consistent allocation to bonds. I am very curious to see how the min corr strategy performs in a rising rate environment vs these benchmarks
Total returns are used. The backtest rebalances monthly back to equal weights even if there are no changes. That is one of the primary differences between a test and real-world portfolio - certain assumptions need to be made in tests.
ETF Replay Portfolio Review: June 2013 [View article]
Shareholder Yield Strategy: Beyond Dividends [View article]
New Ivy Portfolio Tool [View instapost]
ETFReplay Portfolio For May [View article]
ETFReplay Portfolio For May [View article]
ETFReplay Portfolio For May [View article]
All Star Battle: Graham Vs. Piotroski [View article]
Absolute Momentum Portfolio Strategies [View article]
Absolute Momentum Portfolio Strategies [View article]
Absolute Momentum Portfolio Strategies [View article]
All-Weather ETF Portfolio: Minimum Correlation Allocation [View article]
All-Weather ETF Portfolio: Minimum Correlation Allocation [View article]
Dual Momentum Investing [View article]
High Yield Dividend Champion Portfolio For February 2013 [View article]
All-Weather ETF Portfolio [View article]