Strategies Backtest

On Vesak Day, 7 May 2020, I kinda reached my own enlightenment "aha" moment when I backtested 2 strategies on STI ETF:

1) Dollar Cost Averaging (DCA)
~ assuming fixed $1000 invested at open of Tuesday either on or after 12th of each month
~ distributions received are either re-invested the next month's Tuesday in one lump sum or "smoothed" over a few months

2) Weekly RSI 31 mark invest half
~ Fixed $1000 accumulated every month along with distributions into capital fund
~ Invest half of capital when 2 conditions are fulfilled:
i) Weekly RSI hit 31 or below
ii) Weekly RSI tick up
once both conditions are fulfilled, half of the capital accumulated will be bought at the coming week's open

If you are interested, you can click here to see the excel I used to manually backtested.

Both strategies were applied over a 7 years and 4months from 2013 January to 2020 April, when most indices recovered somewhat near half of the losses.

In case you are wondering, why 2013, its the period when I started tracking my trades and returns and posted them in this blog. I was reflecting had I used other strategies, what might happen? Is there a better way and so on. After all, even Warren Buffett says if you unsure, just buy index ETF "a portion of United States".

I was pondering what if I bought "a portion of Singapore" instead of my various trading strategies over those years.

The Enlightenment "aha" moment
The results are as follows:
Strategy 2 "Weekly RSI 31 mark invest half" is more superior, it ended on April 2020 (STI ETF price at $2.605) with portfolio of 25,600 units worth of $88,932.08 (including $22,244.08 cash) against injected capital of $84,000. A return of 5.9% over 7years & 4months or about 0.8%pa.

Both strategies reinvest distributions received.

Strategy 1 "Dollar Cost Averaging" with "smoothing" ie. spreading out reinvestment of distributions received instead of all of it in the following month proves slightly better.

"lump sum distribution reinvested" Portfolio of 31,000 units worth was $81,051.30 including $296.30 cash against $84,000 capital netting -3.5% or about -0.50%pa

"smoothed distribution reinvested" Portfolio of 30,800 units worth $81,295.20 including $1,061.20 cash against $84,000 capital netting -3.2% slightly better, and about -0.46%pa.

Assuming one continue to hold STI ETF after market recovers eventually, it should return around 3.5-4.6% annually.

One thing to note, interests earned are excluded, to focus solely on how the strategies match up with my current portfolio and strategies.

Second factor that work against the DCA approach was the $30 of commission fee assuming 0.28% brokerage fee+other CDP charges etc.which totalled a whopping $2,700 over 7years+. While the Weekly RSI approach based on 0.29% as each invested amount is much higher than $1000 each month; the brokerage costs was only $206.42 with only 4 purchases that match the 2 conditions.


The dismal return is due to the fact that STI give more return via dividend and was almost flat in capital gain as compared to other indices like S&P500, Dow or Nasdaq which have been rallying strongly.


Though STI ETF approach is more safe against individual companies which my blog have shown, the likes of Midas, Foreland that become suspended despite their accounting value worth, shows diversification does helps in minimizing catastrophic losses since STI ETF routinely rebalance itself with the most representative 30 companies on SGX.

I think its enough for today, looking at long term using weekly indicators help during times of uncertainty, though one should note weekly RSI did not get the absolute bottom and can be off by 5-10% from the low. Nonetheless, over the longer term, it may be a good idea to mix or buy more during significant pull backs to oversold zone as indicated by weekly RSI.

Good luck to your trading & investment & stay safe during the circuit breaker in Singapore!