In the Word file attached to Part 10-4 of the Feynman Portfolio Study I included some comments on “benchmarks” – which ones might be used (common index, appropriate Fund, customized index e.g. ITA Index…etc), what to expect when comparing portfolio performance to the selected “benchmark” and whether we should expect to “beat the index/market/benchmark”. For some reason I lost some of the text that I had written, so I will try to remember what I said and repeat it in this Post. I have also included the results of a new backtest to try to clarify the points I would like to emphasize.
In the Word file I noted that if our objective was simply to “match” our selected benchmark then the simplest way to do this was to simply buy the corresponding matching asset e.g. the VTI ETF if we wanted to match the performance of the VTSMX Fund or 4 ETFs representing the allocation to US and International Equities and Bonds if we wanted to match the performance of the VTTVX Fund (or simply buy the Fund itself).
The portion of text that I lost addressed the question as to how we might expect to “beat the index/market/benchmark” if we did not wish/have time to go through some of the more complex processes of momentum analysis, cluster analysis, MPT/MVO Optimization or even setting up a Strategic Asset Allocation (SAA) Plan.
In such a situation, the simplest solution would be to buy the “benchmark” equivalent (e.g. VTI for VTSMX) when price crossed above a chosen EMA (e.g. 195-day EMA -ITARR method) and sell when price dropped below the EMA.
SHY momentum would probably be a better filter, but this would necessitate running the Ranking spreadsheet.
A slightly more sophisticated strategy (although not difficult or time consuming) would be to use Lowell’s VTI, TLT, SHY strategy – again requiring the use of the Ranking SS but not too demanding on time.
A Ranking spreadsheet is set up with SHY, VTI and TLT as the only assets to be ranked. The “Rules” for the strategy are then:
- If SHY is the top ranked asset, 100% is invested in SHY (0% in VTI and TLT);
- If VTI or TLT is the top ranked asset, 75% is invested in the top ranked asset (VTI or TLT);
- Unless SHY is the top ranked asset, 25% is invested in the second ranked asset (VTI, TLT or SHY).
The following Figure shows the performance of the VTI_TLT_SHY Portfolio, adjusted monthly, over the same 7 year period covered in the Feynman Study:
As can be seen from the above figure, the VTI_TLT_SHY portfolio performs better than the “benchmark” VTSMX and VTTVX Funds over the 7 year time period, although it lags behind the more sophisticated Feynman portolios over the same period.
- CAGR 10.73%
- Volatility 10.40%
- Sharpe 1.03
However, it should be noted that from the low of Feb 09 to today July 14, VTSMX has increased from ~$50,000 to ~150,000, a ~200% increase in value, whereas the VTI_TLT_SHY portfolio has increased from ~$100,000 to ~$200,00, “only” a 100% increase. For comparison, the Feynman EW portfolio increases from ~ $105,000 to ~280,000 (~170%) over the same period.
It may be more appropriate to compare with the VTTVX benchmark than has increased ~133% ($60,000 to $140,000) over the same period since VTTVX is a more diversified benchmark comprising 70% (US and International) equities and 30% (US and International) bonds.
Either way, the significant take-away from this analysis is that the greatest benefit to portfolio performance comes from active Risk Management – although the VTI_TLT_SHY portfolio does NOTHING over the first 2 years it avoids the 50% loss experienced by “Buy and Hold” benchmarks.