In part-1, we covered the rules for risk parity, trend following and momentum filters. Then we covered the performance profile of applying risk parity and trend following in asset allocation.
In part-2 we covered the performance profile of applying risk parity & trend following filters together in asset allocation. Then we covered momentum & trend following filters performance when applied to broad asset classes. Then we covered the filters performance when applied to sub-components of broad asset classes.
One issue that came up when applying momentum & trend following filters as a combo overlay was the increased volatility and draw downs. So for the next test, the authors use volatility adjusted momentum ranking. Volatility adjusted momentum is calculated by dividing the prior 12 month return of the asset by the realized volatility over the same period. Table-8 provides the performance details. Gist is volatility adjusted momentum ranking shows some improvement over unadjusted momentum ranking but not much improvement. Following table provides the performance details.
Finally the authors take a flexible asset allocation approach i.e., instead of defining broad asset classes, they just pool all the sub-components and rank them by momentum and trend-following rules. Then they let market decide into which asset classes one should invest.
The biggest advantage of this is it removes any prejudice from the practitioner on the portfolio composition. In other words, one doesn't need to do any judgement call on whether bonds yields are too low to provide any long term value or whether commodities are poor bet in future etc.
Please let me know if you find this series useful and also any interesting papers you come across. Wish you all good health and good trading!
0 comments:
Post a Comment