Momentum Investing for the Australian Market

Amateur trading mistakesIs longer time frame Momentum investing suitable for the Australian Stock Market or more specifically could we run a monthly Australian momentum strategy? If you missed it, we reviewed a US momentum strategy HERE.

Let’s test it.

We’ll use the All Ordinaries Index including all historical constituents and delisted stocks thanks to Premium Data. Commission is set at 0.08% or minimum $6.

In this first test we’ll compare like-for-like as per the discussion in the US version and Andreas Clenow’s Stocks on the Move parameters, specifically:

momentum lookback = 90 days
stock filter = 100 days
index filter = 200 days.

1999 – 2014        XAOA     Momentum
Annual Return      8.7%       20.5%
Max Drawdown   -51.4%      -28.7%

Some very strong data there. Significant return outperformance and whilst the drawdown is higher than what was offered by US markets, it’s still considerably better than a buy and hold strategy.

In this second test we’ll use the nominal number of trading days on the NYSE across all three variables:

momentum lookback = 252 days
stock filter = 252 days
index filter = 252 days.

1999 – 2014         XAOA    Momentum
Annual Return       8.7%        24.6%
Max Drawdown    -51.4%     -30.2%

Still strong results. We can probably assume the higher drawdown could be accountable by the broader range of smaller cap stocks contained in the All Ordinaries vs those contained in the S&P 500. A 30% drawdown is probably at the upper end of most peoples tolerance, even if it’s still well below buy and hold.

Lastly, let’s use 5-years of out-of-sample data, namely 4-years prior 1999 and 1-year post 2014.

1995 – 2015         XAOA    In-Sample    Out-Sample
Annual Return       9.3%      24.6%          24.5%
Max Drawdown    -51.4%    -30.2%        -30.2%

Some benefits to digest:

  1. Hard to argue it’s robustness and because of this we can probably squeeze some more juice out of it by finding better parameters through optimisation.
  2. Let’s also remember we’re only making trades once a month. Very low workload for some very good results.
  3. There is no Sample Bias with this strategy (unlike a trend following strategy). In other words there is only ever one set of trades so no ‘picking or choosing’ between trades is required.

The strategy fared well in 2008 losing -11.8% for the year. This was attributable to maintaining a defensive stance and reinforces our long held belief, previously discussed here, that at certain times no trade is the best trade. Attempting to time the best possible entry into a strategy is a pointless exercise due the short term randomness of market returns.

It’s important to maintain a long term application of any strategy so it’s edge can be exploited over full market cycles.