Last week I outlined a simple Rate of Change [ROC] trading system discussed at OASIS 2008. This week a variation that incorporates options is introduced. The approach updates the stock list to eight optionable securities in different sectors with small bid-ask spreads to minimize slippage costs.
System Description
As a quick review, the base system identified at OASIS used an ROC crossing above or below its simple moving average [SMA], with two different ROC settings for entry and exit. Table 1 provides details for the initial back-test that was performed.
| System Details |
Entry | 34-dy ROC cross above 13-dy SMA |
Exit | 21-dy ROC cross below 8-dy SMA |
Start Value | $20,000 |
Allocation | 50% |
Brokerage | $10 per trade ($20 roundtrip) |
Period | 7/1/1999 – 6/30/2005 (6 years) |
Stock List: | 39 Nasdaq 100 stocks |
Table 1: Base ROC System Details
Results for this system are marginal from a profit standpoint, but the real appeal is in its stability. The ROC base serves as a good starting point for improving results with filters. As mentioned at OASIS, a consistent improvement in performance was realized for timed exits of 15 or 20 days. Timed exits of 5 and 10 days led to consistently poor results. This makes sense since the average time in winning trades was 19 days and the average time in losing trades was 10 days.
This is good news for option traders since entering a trade on the day a momentum signal is generated may lead to increased costs due to an increase in the option’s implied volatility [IV]. By delaying entry with the base ROC system, the option trader may capture a modestly lower price for the underlying stock along with lower IV for the option selected. The IV impact can be further minimized by selecting an in-the-money [ITM] option. By also incorporating a timed exit, the trader can readily identify an appropriate expiration month for the trade.
As a quick reminder, successful trading focuses on minimizing risk. When using a system to trade, risk is often minimized by incorporating stops. Stops were tested with this system; however, they consistently produced unfavorable results. This can happen with trend systems. If a stop cannot be identified, then the system’s money management approach must assume a potential 100% loss for each position. This means that the amount allocated for each trade must be within your personal risk management parameters for maximum risk per trade. In addition, the potential for consecutive losses must also be considered. Once again, options lend themselves nicely to such constraints since they represent a leveraged trade with intrinsic and extrinsic value.
Table 2 provides the updated ROC system for the ROC system using delayed entry and exits on a revised short-list. In this instance a 50% allocation of the total account value was used rather than a 50% allocation of the starting value, which would peg the trade amount to $5,000. Unfortunately my ProfitSource (PS) laptop is currently out of commission so I’m unable to run it again with that correct allocation. This is only a problem if the system runs out of cash to trade, which doesn’t occur with this system. The profit/loss percentages can still be used.
| Revised System Details |
Entry | 5 calendar* days after the 34-dy ROC crosses above its 13-dy SMA |
Exit | 16 calendar* days after the 21-dy ROC crosses below its 8-dy SMA |
Start Value | $10,000 |
Allocation | 50% of account |
Brokerage | $10 per trade ($20 roundtrip) |
Period | 7/1/1999 – 6/30/2005 (6 years) |
Stock List: | AMGN, GOOG, GS, IVGN, NEM, NTAP, SBUX, XOM |
Table 2: Delayed Entry & Exit ROC System for Revised Stock List
* Since the timed entries and exits used in PS incorporates calendar days instead of trading days, the use of 5 days corresponds to approximately 3 trading days and 16 calendar days corresponds to approximately 12 trading days.
The system generated 100 trades for the period, with an average profit of 4.5% and a median profit of 3%. More profits were generated during bullish periods for the market; however, profits and stability were maintained during bearish periods. Using securities that are not positively correlated helped, but most of the eight securities generated net profits during the 2000-2002 bear when tested individually.
Sixty-seven percent of the trades were profitable which is unusually high compared to other variations on the base system. 4 of the 100 (4%) trades had a Maximum Favorable Excursion (MFE) of 0 versus the 40% MFE in the base system. The greatest Maximum Adverse Excursion (MAE) % for the system was 16% and the largest loss was 14%. The maximum cumulative loss for the system using consecutive losing trades was 23%.
The end of period account value was $52, 359.24, which is significantly reduced when using $5,000 per trade rather than 50% of the account value. There’s no way I’d create a position size of $25,000 without a stop, so just remember to be realistic about a system you would actually put into action. The return for SPY, the S&P Depository Receipt for the S&P 500, had a net return of -14.1% for the same period the system was active (7/7/99 – 6/23/2005).
Figure 1 provides a scatter plot of the profit per trade for each trade (total of 100).
Figure 1: Scatter Plot of Profits/Losses per Trade for revised System
Using closing prices for August 8, 2008, the eight securities have spreads that represent 5% or less of the option ask price when evaluating next month calls with more than 30 days to expiration that are at or near the money. As an example, AMGN closed at 64.21 and the Sep 62.5 calls had 36 days to expiration. The option spread was 0.15 or 3.6% of the ask price (4.00 x 4.15).
Next week results from a forward test of the revised ROC system will be provided, along with the results for option trades used in place of stock trades.
To access other articles written by Clare White, please click here.
Clare White, CMT
Contributing Writer and Options Strategist
Optionetics.com ~ Your Options Education Site
Questions for Clare? Visit the Optionetics.com Discussion Board
