Sometimes, the good stuff needs to wait until after a seminar is over.
As speakers in BIGSCribe events are also investors, we are also part of the audience when someone else is speaking.
Kenny Loh or Marubozu gave a fantastic presentation on REITs investing and runs a course here. Here are the results of my back-testing to refine my own REIT investing strategy using the insights I learned from Kenny Loh's Three Musketters approach to REIT investing. Paying customers would already have some sort of quick tutorial on what semivariance is from my presentation.
a) Baseline - buying all REITs at one go
If you buy all 41 REITs in equal proportions, your returns would have been 8.5% with a semivariance of 13.67% for the past 10 years. This is our baseline and I recommend that every investor who might not want to go too deep into screening should just buy all the REITS in SGX in equal proportions.
b) Choosing REITs with the highest Yield
As I have spoken in my own presentation, buying half of the higher yielding universe of REITs can outperform the strategy of buying all the REITs in the SGX universe. Last time I backtested 9.64% with a higher semivariance of 14.25%.
c) Choosing REITs with the lowest Gearing
Kenny spoke about looking for REITs with a lower gearing. I backtest a strategy that buys half of the REITs in SGX with a lowest debt to equity ratio. Once again, I was able to outperform at 9.56% with semivariance of 13.98%.
d) Choosing REITS with the lowest Price to Net Asset Value
Kenny spoke about being careful when looking for REITs with a high net asset value. I backtest a strategy that buys half of the REITs in SGX with a lowest pice to book ratio. This time I underperformed at 7.28% with semivariance of 15.08%.
Attempting to buy a dollar worth of real estate with 99 cents actually backfires on the investor with lower returns and higher risk.
e) Super-duper REIT screening strategy
So thanks to Kenny, there are at least two working strategies. Find REITs with a high yield and low gearing. I combined both screens, searching for the top 50% highest yielding REITs and then within that set, short-listing 50% of those with the lowest gearing.
This time I had a winning strategy in my hands. A final return of 13.16% with a semivariance of 14.49%. It has a fairly high Sharpe ratio of 0.54. 1 in 40 years, you may lose about 16% of your portfolio value, making this something which may be amenable to 200% leverage.
What is the moral of the story ?
When investors get together and mutually present seminars, our insights are silo-ed and we might not be able to extract the maximum benefit if we stick to our own investing approach. Even my 6-8% strategy returned only 10%.
Because I always make sure that I follow up on my learnings from other speakers, the blogosphere can benefit from a much sharper insight that combines the investment ideas of several speakers.
[ Note : The Singapore REITs universe is small, applying a screen to choose a quarter of sticks in the universe will only yield about 8-11 stocks. A diversified investor will need a few different strategies to build a portfolio that can withstand the test of the time. ]
Thank you for sharing your back testing results. Would you mind to share which are the reits based on your 50% highest yield followed by 50% lowest gearing screening? I assumed when you used the data 10years ago for screening.
ReplyDeleteI did not want to share it on purpose because I was concerned that too many would just follow the screen and not do their own research.
ReplyDeleteI also want to ensure that folks read all the other stuff in the local blogosphere on overfitting before proceeding to screen for their own stocks.
Would you be able to share how you did your back-testing? What tool do you use so that I can do my own testing as well.
ReplyDeleteHi Chris,
ReplyDeleteI have been using your methodology for the past year with great success. >10% IRR in the last 6 months (although I am sure big portion of it is due to general REIT market bubble).
As I am rebalancing my portfolio now, i realised I have a couple of questions on using this method.
1. Did you use Debt/Equity ration or Debt/Asset ratio for the Gearing Ratio filter? I tried both ways and got 50% common results.
2. Did you strictly limit to REITS or included all REITS + Trusts (e.g. Netlink Trust, Asia Pay TV Trust, Hutchinson Port Trust)?
Thanks in advance for sharing.
It is probably not wise to continue to use gearing ratio as a filter.
ReplyDeleteAlso I do not use business trusts in my REITs filters.