(published in the VantagePoint Strategies Newsletter)

In Part 1 of this article I described using VantagePoint’s Predicted Neural Index indicator as a simple trade entry signal, and presented some backtesting results using that approach with the USD/CAD forex pair.  In this article I will add another VantagePoint indicator to the approach and show how it can greatly improve trading performance.

The Predicted Neural Index

The VantagePoint Intermarket Analysis software provides numerous predictive indicators a trader can use individually or in combination to inform trading decisions.  The Predicted Neural Index, PIndex, is one key indicator that can be used.  It displays a value of ‘1’ or ‘0’ to indicate expected short term market direction - a value of ‘1’ indicates prices are expected to rise, and a value of ‘0’ indicates prices are expected to decline.

VantagePoint determines the PIndex value by comparing the value of two 3-day simple moving averages – the 3SMA of typical prices on the current day (“analysis day” or day of interest), and the predicted 3SMA of typical prices two trading days in the future.  VantagePoint uses neural networks and intermarket analysis to develop this prediction. 

Review – Basic Approach

In Part 1, I backtested performance of a method using changes in the PIndex value for the USD/CAD as a trade entry signal.  The simple rules used in the analysis were: 

(1)    When PIndex changed value, I entered a trade – ‘long’ if the new value was ‘1’ and ‘short’ if the new value was ‘0’;

(2)    I assumed a stop loss was placed 20 pips below the low of the analysis day for long trades, and 20 pips above the analysis day high for short trades;

(3)    I exited a trade when the stop loss was reached, a profit target of 100 pips was reached, or a profitable daily closing price was reached;

(4)    I also exited a trade if PIndex changed value - in this case I ‘stopped and reversed’, to be trading in the direction of the latest VantagePoint prediction. 

Backtesting this approach for 2010 (thru early May) showed a win rate of 71%, which was an encouraging result for such a simple method; unfortunately the method suffered several losses from that point forward.  In further reviewing the backtesting results, I observed that increasing the stop loss level slightly, to 30 pips rather than 20 pips above/below the analysis day high/low, could have a positive effect on results, and so I made that change in the analysis.  Unfortunately even this adjustment could not prevent the method from experiencing several whipsaw losses in late May and early June, and achieving a net loss for the year as of June 15. 

Adding a Second Opinion  

Since VantagePoint uses multiple neural networks to produce values for its various predictive indicators, it can be useful and powerful to compare the ‘opinion’ or behavior of more than one indicator and seek confirmation between them.  One such indicator we can use in conjunction with the PIndex is VantagePoint’s ‘short term crossover’.

This indicator consists of the 5SMA of closing prices and the 2EMA of typical prices (i.e., average of the day’s high, low and closing price) predicted 1 day in advance.  One way these moving averages can be used is to enter a trade when a new cross of the moving averages occurs, entering in the direction the EMA crosses the SMA.

For example, if the P2EMA crosses the 5SMA upwards from below, a long trade is entered; and if the P2EMA crosses the 5SMA downwards from above, a short trade is entered.  As shown in the chart below, using the moving average crosses as trade entry signals can be a viable trading approach all on its own (the green ellipses indicate winning trades and the red indicates a losing trade).  The blue arrows on the chart indicate trade entry direction, with ‘up’ indicating a long trade, and ‘down’ indicating a short trade, corresponding with the moving average crossover direction.

 By adapting the earlier trading rules to this entry signal alone (i.e., without PIndex) I was able to backtest this approach for 2010 (thru June 15) and achieve very positive results.  Twenty one trade entries were signaled, with 17 wins and 4 losses, for a win rate of 81% and a total of 646 pips.  Furthermore, some spreadsheet analysis of the trade details showed that risking just 5% of current account balance on each trade would’ve achieved a total 29% gain in equity thus far in 2010.

 Combining the Indicators

The question remaining to be answered is, how will things go if we combine the signals provided by the PIndex and the short term crossover to determine trade entries?  Specifically, the rules to be backtested are: 

(1)    When PIndex changes value, and a fresh short term crossover occurs in the same direction, a trade is entered

-         ‘Long’ if the new PIndex value is ‘1’ and the P2EMA crossed the 5SMA upwards on the day of or the day following the PIndex change;

-         ‘Short’ if the new PIndex value is ‘0’ and the P2EMA crossed the 5SMA downwards on the day of or the day following the PIndex change;

(2)    I assumed a stop loss was placed 30 pips below the low of the analysis day for long trades, and 30 pips above the analysis day high for short trades;

(3)    I exited a trade when the stop loss was reached, a profit target of 100 pips was reached, or a profitable daily closing price was reached;

(4)    I also exited a trade if an opposite direction entry signal occurred - in this case I ‘stopped and reversed’, to enter a trade in the direction of the latest VantagePoint prediction. 

The combination of these predictive indicators can be very beneficial.  At times, the PIndex may ‘chatter’ (rotate back and forth from 1 to 0 and back again) as the market moves sideways or forms a top or a bottom; and at times a short term crossover in the moving averages may occur although the day’s action causing the change may be only a temporary jab or due to a very short lived situation in the financial markets.  But when used in combination, some of these less desirable entry signals can be eliminated.

For example, on April 29th, marked ‘1’ on the chart below, the PIndex changed value from ‘1’ to ‘0’, indicating a possible downward move in future prices;  however, the P2EMA did not cross the 5SMA, and hence a trade was not entered.

On May 3^rd, marked ‘2’ on the chart below, the P2EMA crossed the 5SMA downward from above, indicating a possible move downward in prices;  however, the PIndex remained ‘1’ and hence a trade was not entered.  In both these cases, the combination of PIndex and the short term crossover indicators led to correct ‘no trade’ decisions.

This chart shows only a 1 month time period in 2010, during which 3 trade entries were signaled and all 3 were winners (circled in green).  Thus far in 2010 (thru June 15th), this “combination quick strike method” has signaled 19 trade entries, with 16 winners and only 3 losers, for a win rate of 84%.  A total of 644 pips were generated, and analyzing the trade details showed that risking 5% of account balance on each trade would’ve resulted in a 32% gain in equity thus far in 2010.  The plot below shows the growth in net pips and account balance thru the year, starting with a $1,000 balance.  

I don’t know about you, but I look forward to trading with these signals in the weeks ahead!  Cheers all, Good Trading.