As many of you will know we started an experiment last month on the use of the Relative Strength Index to buy undervalued stocks.
The theory goes that an RSI approaching 30 (or below) is oversold, and the stock may now start an uptrend, and an RSI approaching 70 (or above) is overbought, potentially causing the asset to start a downtrend.
As of October 25 2014 there were 703 stocks that are approaching an RSI of 30. For the purpose of this experiment we are going to class 'approaching an RSI of 30' as any stock below an RSI of 40 on 14 day basis.
For this investigation we looked at the
what if scenario for portfolio performance if buying all stocks
approaching an RSI of 30, all stocks
below an RSI of 30, as well as a
specially selected group of stocks approaching an RSI of 30 that satisfy a few fundamental factors. We also sneaked in a fourth group of 20 stocks that were selected at complete random to test the validity of the results more so.
The
special group of stocks, shown in the image below, were picked as they satisfied certain criterion necessary to be included in this particular portfolio.
- Return on Equity > 20%
- Debt to Equity < 80%
- PEGY Ratio < 2
- Relative Strength Index < 40
Well, the results are in and they are somewhat interesting. For the record, we plan to repeat this experiment a few times to account for different market conditions and to test reliability, as we all know by now that past results do not guarantee future results.
The results:
So, as you can see disappointingly the random selection ended up the victor, outperforming all selections and the market during these 20 trading days. But that's not the whole story. There seems be a case here for a reduction in the holding period for each of the groups of stocks. 9 days turned out to be the optimal holding period during this experiment.
|
RSI Special |
RSI < 30 |
RSI < 40 |
Random |
Market |
| Day 9 |
6.07% |
1.95% |
3.86% |
3.22% |
1.19% |
As you can see above, as of Day 9 every single group of stocks outperformed the market, and by a reasonable margin. The decline that follows shortly after could show that potential RSI gains are ephemeral in nature and do not necessarily end up in overbought territory (RSI >70) before they decline once again. We should note that we were especially pleased with performance of the specially selected group of stocks at Day 9. Picking a group of downbeat, but reasonably sound, stocks rather than a blanket selection of all stocks approaching an RSI of 30 seems logical and appears valid.
We'll let you know how things go the next time we test this strategy. Feel free to try the strategy yourself and let us know how you get on.
- J
