Please Note: Blog posts are not selected, edited or screened by Seeking Alpha editors.

Myth Or Truth: Is The Golden Cross A Shiny Investment Approach? - An Empirical Investigation

|Includes: iShares Core S&P 500 ETF (IVV), SPY

You probably have read a lot about the "Golden Cross" on different financial media and maybe you even thought of following such an investment strategy. The "Golden Cross" is a technical pattern which occurs when the 50 day moving average of a specific underlying security crosses above its respective 200 day moving average.

The claim is that this signal quantifies the improvement in the trend of the underlying security and will lead to a significant uptrend. Since a moving average is an indicator that measures the average movement of an asset price over time, it is a confirmation of upward momentum when a short-term moving average crosses above a longer-term moving average and vise versa.

Conversely, when the 50 day moving average crosses below the 200 day moving average, it is known as the "Death Cross", and is considered as an exit point. This technical pattern assumedly signals a significant downturn.

In the literature (Joseph Granville, 1960) it is sometimes argued that a "Golden Cross" should also have a rising 50 as well as a rising 200 day moving average. For our analysis we will focus only on the "simple" moving average crossover since this version is more widespread among the investing community.

In general there are two common ways to trade such a technical pattern:

  • The first approach (Long/Cash) is to trade only the bullish signals ("Golden Cross"), since not every investor is able to enter a short-position in the specific underlying of interest, while during a so-called "Death Cross", investors just take no investments (NASDAQ:CASH) at all.
  • The second approach (Long/Short) is to trade the bullish- as well as the bearish signals, in taking a specific long as well as short position within the specific market.

Methodology:

For our analysis we are using the S&P 500 (NYSEARCA:IVV) as underlying market.

  • All used data is split and dividend adjusted. In addition, we have taken 2 days slippage into account, but we have excluded any taxes and trading expenses.
  • The analyzing period was from 10/18/1950 until 04/30/2012
  • In the Long/Cash version, no investment is being made during that time a "Death Cross" is predominant.
  • Both, the Long/Cash as well as the Long/Short strategy will be compared with a simple "Buy & Hold" for the S&P 500.

Results

Trade Analysis:

From 10/18/1950 until 04/30/2012 the Long/Cash strategy generated 63 trades. Assuming zero transaction costs, this technical pattern would have been profitable with a mean and median profit per trade of 17% and 12.9% respectively. In total, 78.1 percent of all trades would have generated a positive return. The best trade would have produced a profit of 124% while the worst trade was down almost 10%. If we would treat the best trade as an outlier and remove that observation from the sample, the mean and median profit per trade would have been still above 10%.

The results for the Long/Short strategy for the same time period are not as promising compared to the Long/Cash strategy. In total, this strategy generated 129 trades, as investors have additionally entered a short-position, if a Death Cross signal appears.

Again, assuming zero transaction costs, this strategy would have been profitable with a mean and median profit per trade of 9.24% and 0.40% percent respectively. All in all, trading the Death Cross as well, this would have lowered the mean and median profit compared to the Long/Cash strategy, by 7.76% and 12.5%, respectively. In total, 55.4 percent of all trades would have generated a positive return.

Trade Statistic
Table 1: Trade Overview

Performance Analysis:

Table 2 summarizes the results for both Long/Cash and Long/Short. It also displays the relevant values for the naive buy-and-hold strategy for easy comparison.

The results of back-testing the Long/Cash- and Long/Short strategy illustrate that on a day-by-day basis, those strategies slightly outperformed a naive buy-and-hold-portfolio but we have taken zero transaction costs into account.

The Long/Cash portfolio has the lowest volatility, mainly due to the cash investment. Nevertheless this portfolio has the second highest annualized return and due to its low volatility it has therefore the best Sharpe ratio among all three portfolios.

The main reason, why the Long/Short portfolio has outperformed the naive buy & hold- as well as the Long/Cash portfolio in terms of annualized returns, is the fact that this strategy has performed very well during the past long lasting bear markets (Table 5).

Performance/Risk Ratios
Table 2: Performance Analysis

Calendar Returns Analysis:

In total, only in 34.9 percent of all observations regarding the yearly results, the Long/Cash as well as the Long/Short version outperformed a Buy & Hold strategy.

If we have a closer look on the dispersion of the yearly results (Table 6), we can see that the Long/Short Strategy gained a whopping 69.80 percent on a yearly basis, in 2008, compared to only 45% for a classical Buy & Hold Strategy in the year 1954.

In terms of maximum yearly loss, both the Long/Cash as well as the Long/Short Strategy are clearly outperforming a typical Buy & Hold portfolio by at least 15%.

Calendar Returns of the S&P 500 since November 1950 in detail:

Calendar Returns S&P 500

Table 3: Calendar Returns of the S&P 500

Calendar Returns of the Long/Cash Strategy since November 1950 in detail:

Calendar Returns Long/Cash
Table 4: Calendar Returns of the Long/Cash strategy

Calendar Returns of the Long/Short Strategy since November 1950 in detail:

Calendar Returns Long/Short Strategy

Table 5: Calendar Returns of the Long/Short strategy

Dispersion of Yearly Results since November 1950 in detail:

Dispersion of Yearly Results
Table 6: Dispersion of Calendar Year Results

Drawdown Analysis

If we have a closer look at the historical five largest drawdowns, we can see that the main benefit of the Long/Cash- as well as the Long/Short strategy is their downside protection.

Table 8 and Table 9 show how those two strategies would have significantly reduced the maximum drawdowns compared to a simple buy & hold portfolio (Table 7).

The maximum draw down of the Long/Cash portfolio was only -33.2 percent compared to the naive portfolio which has lost -56.8 percent during the financial crisis. Even the more risky Long/Short strategy has a lower maximum drawdown than the buy & hold portfolio. Nevertheless, the recovery period is the longest when investors follow the Long/Short strategy. The Long/Cash strategy has on average the shortest recovery period, if we have a look at the five largest drawdowns.

Largest Drawdown of the S&P 500 since November 1950 in detail:

Largest Drawdowns S&P 500

Table 7: Largest Drawdowns S&P 500

Largest Drawdown of the Long/Cash strategy since November 1950 in detail:

Largest Drawdowns Long/Cash

Table 8: Largest Drawdowns Long/Cash strategy

Largest Drawdown of the Long/Cash strategy since November 1950 in detail:

Largest Drawdowns Long/Short

Table 9: Largest Drawdowns Long/Short strategy

Conclusions

The testing results show that both, the Long/Cash and the Long/Short strategy have been very reliable since 1950. In the past, 78.1% and 55.4% of the trades for the Long/Cash and the Long/Short strategy respectively, have been winners.

Exiting on a Death Cross would have kept investors out of most ugly bear market declines in the past. So if you believe that you are in an environment where lengthy declines of more than 30% are going to be the order of the day, following the Long/Short- or Long/Cash strategy instead of a Buy & Hold approach, can be useful.

Upon a Death Cross signal the market shows general weakness to perform on the upside. This weakness can be difficult to short, if the there are no long-lasting periods of weaknesses. Moving into cash or selling short during this time may prevent you from experiencing significant drawdowns compared to a simple buy & hold strategy.

However, as this method confirms an existing trend rather than predicting one, there is a chance that investors face a late-in and late-out problematic. Therefore in only 34.9 percent of all observations, those two strategies have outperformed a buy & hold portfolio in the long-run.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.