In the article Market Timing: 3 simple indicators that work I stated that three market timing indicators of different kinds may help investors to improve their risk adjusted performance, mainly by cutting the risk factor. A few comments and private messages wondered if these indicators could be combined. This is my answer.
For convenience I will repeat the definitions of the signals:
- "Golden Cross" is bullish when the 50-day simple moving average of SPY is above or equal to its 200-day simple moving average. Else, it is bearish (then called "Death Cross"). It is a technical indicator.
- "EPS Estimate" is bullish when the S&P 500 current year EPS estimate is above or equal to its value three months ago. Else, it is bearish. It is an aggregate fundamental indicator.
- "Unemployment" is bullish when the U.S. unemployment rate is below or equal to its value three months ago. Else, it is bearish. It is a macro-economic indicator.
There are different manners of combining them. The next table shows the results for SPY using the most obvious ones (period: 1/2/1999 - 1/17/2014). Dividends are included.
Buy and Hold
One is Bullish
Two are Bullish
Three are Bullish
The "One is bullish" and "Two are bullish" versions give the same annual return, the second one delivering the lowest volatility and drawdown.
"Three are bullish" gives a return barely above "Buy-and-Hold", but it is the lowest risk tactics.
Here is the equity curve for "Two are Bullish" with SPY:
It is also possible to test variations: using the indicators two by two (3 possibilities), with an error margin (for example considering that an indicator is bullish until 0.5% or 1% below or above the limit). The possible combinations can quickly grow up. An in-depth study would be too long here, readers who want to do it can find instructions here, then ask me in private message the code of indicators. I will be happy to give it for free.
These three indicators, alone or combined, may cut drawdowns and boost the market's long-term total return. The best and simplest way to use them in a recent past was to stay in the market when two are bullish, and go to cash when two are bearish. Simulations are performed on a long period, but a relatively small set of signals: I wouldn't conclude that a solution is much better that others. At the time I write this (1/23/2014 using data on close the day before), the three indicators are bullish.