Harvesting Momentum: Let's Kick Tires With AmiBroker [ Part III - Final ]

The previous posts in this series covered a basic Tactical Asset Allocation approach with with signals taken from the daily and the weekly time frames. This final post will be dedicated to portfolio rotation based on monthly and quarterly time frame. Only two asset class universes will be covered. Multi asset class universes are going to be the subject of a future post.

Taking the monthly time frame into consideration basically means rotation decisions are made once a month, after the close of the month. For the quarterly time frames this comes to only 4 re-balancing opportunities per year: at the end of each quarter. However, usually the re-balancing frequency is way lower.

For backtesting once again the source code provided in the opening post will be used, basically allowing the DIY retail investor to engineer his own alpha. Even if you don't have access to thinkorswim or Amibroker, the strategy at hand can be fairly easy implemented using Excel, since momentum is the only metric taken into consideration for re-balancing decisions.

Monthly Periodicity

The typical US stock market year has 252 trading days (= 12 months with 21 days on average). So for the previously used 85 days/17 weeks lookback period, momentum calculated over a 4 month period (= 4 x 21 days) is suitable. Applied to the base pair (SPY-TLT) this 4 month lookback period renders the following results.

Portfolio Equity SPY - TLT (4 months), drawdown periods in red
Drawdown graph SPY - TLT (4 months)
Profit Table SPY - TLT (4 months)
SPY-TLT (4m)    2004 - 2013
Total Profit   265     %
CAR   13.94%
MDD (trades)   -17.13%
MDD (system)   -17.13%
Calmar (0%)   0.81
Sharpe (0%)   0.94
Trades   17    
Winners   82.35%

A quick comparison against the backtester results for this pair on the daily and weekly time frames learns that all monthly metrics are best except for the CAR which is 0.75% better with the daily signals.

While smoothing for the monthly time frame is rather confined, some smoothing can still be applied.
Performance of SPY - TLT sorted on Calmar (CAR/MDD) on x-month momentum after y-month smoothing (CAR > 10%)
The table shows overall results for the 3 month momentum are slightly better. As this setting also matches the quarterly time frame exactly, for the rest of this post the 3m/1q lookback period is default.

Again the replacement of SPY with MDY is a performance booster for the monthly time frame.
Portfolio Equity MDY - TLT (3 months), drawdown periods in red
Drawdown graph MDY - TLT (3 months)
Profit Table MDY - TLT (3 months)
MDY-TLT (3m)    2004 - 2013
Total Profit   538      %
CAR   20.54%
MDD (trades)   -14.40%
MDD (system)   -14.39%
Calmar (0%)   1.43
Sharpe (0%)   0.85
Trades   23     
Winners   78.26%

To conclude the exploration of the Monthly Periodicity for the Stocks - Bonds paired switching let's again take a look at a now familiar mutual fund pair. For stocks: FDVLX and for bonds: VUSTX. For this combination there is more than 27 years of historical data available allowing to start the backtest at the beginning of 1987(!). This means we have the October 19, 1987 Black Monday stock market crash in the backtest. On this day the Dow Jones Industrial Average experienced it's largest daily percentage loss ever: -22.61%.

Portfolio Equity FDVLX - VUSTX (3 months), drawdown periods in red.
Chart is in semi-log scale to demonstrate the horrendous setback in the first year of the backtest
Drawdown graph FDVLX - VUSTX (3 months)
Profit Table FDVLX - VUSTX (3 months)
FDVLX-VUSTX (3m)    1987 - 2013
Total Profit   3,154      %
CAR   13.80%
MDD (trades)   -26.06%
MDD (system)   -26.06%
Calmar (0%)   0.53
Sharpe (0%)   0.76
Trades   64     
Winners   79.69%

Quarterly Periodicity

To bring this series to an end, let's apply the quarterly time frame to the above shown mutual fund pair to get a complete comparison.
Portfolio Equity FDVLX - VUSTX (1 quarter), drawdown periods in red (1987 - 2013, semi-log scale)
Drawdown graph FDVLX - VUSTX (1 quarter)
FDVLX-VUSTX (1q)    1987 - 2013
Total Profit   1,860      %
CAR   11.76%
MDD (trades)   -26.52%
MDD (system)   -26.52%
Calmar (0%)   0.44
Sharpe (0%)   0.67
Trades   49     
Winners   75.51%

The chart below plots the equity curves for the FDVLX - VUSTX paired switching strategies to compare the four investigated time frames against the Buy & Hold equity curve for the Vanguard 500 Index (VFINX) over 1987 - 2013. The chart demonstrates the superior performance of the "simple" paired switching strategies in times of prolonged bear markets. Especially in those adversary conditions the mechanical switch to bonds works as a crash protection keeping drawdowns limited, thus preserving capital and allowing the portfolio to prosper during the bull markets to come.

Portfolio equity comparison for FDVLX-VUSTX over 1987-2013 for daily, weekly, monthly and quarterly rotation.

What's next?

The fulfillment of the announced strategies with multi asset universes will be covered in a (soon to come) future post. So stay tuned.

Performance over 2004 - 2013 for multi asset portfolio with momentum only strategy (semi-log scale)