Forex Trading System- Tools of forex trading

OVERVIEW

In The Forex Chartist Companion(Wiley, 2006), we introduced several innovative
charting techniques and some new technical analysis tools. In this chapter we review
the ones used in this analysis of the major currency pairs.

It is because of the highly chart-intensive nature of the book that we are com-
pelled to provide very precise definitions of our chart time components to avoid any
confusion.

The time frameof a chart is the overall duration that the chart spans. On the left
side of the chart is the starting date and time and on the right side is the ending date and time. Date and time are represented in the conventional MM/DD/YY HH:MM format. The 24-hour (military) time format is used throughout.

The time intervalof a chart is the equally spaced time unit into which the time
frame is divided. In the case of a vertical OHLC bar chart, the time interval is the width along the x-axis of a single OHLC bar, that is, the amount of time elapsed between the opening quote of the OHLC bar and the closing quote of the same vertical OHLC bar.

ACTIVITY

The concept of activity is employed as a means to evaluate the intrinsic characteristics of a specific currency pair and acts as a surrogate tool for trading volume, a statistic not readily available for spot currencies due to the decentralized nature of Forex markets.

Theoretically, activity represents the number of price changes within a given inter-
val of time. Unfortunately, the activity number does not show the size of each order.


MIDRANGE

Another classical statistic based on the highest high and the lowest low is called the midrange and is the midpoint between the two extremes.

RELATIVE RANGE

Whereas the absolute range described earlier is an excellent tool for use in the analysis
of the internal characteristics within a single individual security, relative range is used to compare the characteristics of two or more similar securities.

The denominator is a critical central point (the midrange in this instance) that con-
verts an individualized statistic into a generalized statistic that is ideal for comparing different sets of similar data. Where absolute range is expressed in terms of pips of the quote currency in the currency pair, relative range is expressed as a percentage and acts as a dimensionless index number. It is this characteristic that permits comparisons between different currency pairs.

A relative range chart differs only slightly from the absolute range chart described
earlier in this chapter: The vertical bars in the lower half of the chart are slightly smoothed, and the lower right scale is expressed in percents instead of pips.

Relative range is a measure of relative volatility and can be used to assist the trader in determining which currency pairs to monitor based on the trader’s predilection for the ubiquitous risk/reward factor. Trading pairs with high relative ranges increases the risk factor while also increasing the likelihood of greater profits.

A high relative range does not mean that a currency pair is more actively traded
than other pairs. Instead, it implies that over time the underlying security prices will travel greater distances from a critical statistical point (in this case, the midrange point).

ABSOLUTE MOMENTUM

Standard momentum (one close minus a previous close separated by lagtime units)
generates a stream of data consisting of both positive and negative numbers whose
mean approaches zero in large samples. To rectify this intrinsic mathematical property, it was necessary to use the absolute value of the momentum data streams. That is, all negative numbers are converted to positive numbers.

Thus, when using absolute momentum, we are not concerned about the direction
of the processed data since all absolute momentum values are positive. We are, how-
ever, very interested in the magnitude of the processed data. Extreme values in an
absolute momentum oscillator inform us at what time of day breakouts are most
likely to occur, although we do not know which direction they will take. This is,
nonetheless, valuable information to traders, particularly for those who subscribe to
trend-following techniques.

STANDARD DEVIATION

Of the several methods of calculating the dispersion of a data set from a central point, we prefer to employ the moving standard deviation as the measurement of volatility.

Statistically, standard deviation is defined

where


x = the sample elements (prices)
n = the sample size (number of prices)
Variance= the sample variance (sum of the deviations squared)

Generally the standard deviation increases as a clear price trend begins emerging in
either direction and decreases when lateral congestion originates. A sharp decline in the standard deviation indicates that a price reversal has begun, after which the standard deviation will again increase regardless of the direction of the trend.

COEFFICIENT OF VARIATION

Just as the standard deviation is a measure of absolute dispersion within a single currency pair, the coefficient of variation is a measure of relative dispersion. The standard deviation is always expressed in terms of pips in the quote currency,
such as dollars, francs, pounds, yen, and so on. The coefficient of variation is expressed as a percentage (or dimensionless index number), which makes it an ideal tool for comparing two or more similar data sets.

Traders should not confuse the coefficient of variation with another statistic called
the coefficient of correlation, which measures how closely the estimated values match
the raw data in a specified regression model, such as a linear, parabolic, sinusoidal, or logistic regression. The coefficient of variation is analogous to relative range, described earlier.

COMPOSITE CHARTS

Another important analytic tool that we introduced in The Forex Chartist Companion
is the composite chart, of which we presented two general types: the Time of Day Chart and the Day of Week Chart (defined by their time span, one day or one week respectively).

Composite charts are constructed by averaging one specific statistical category (ac-
tivity, range, or momentum) over a selected time frame. For example, the daily composite chart in Figure 2.9 illustrates the average range on all Wednesdays between 1/1/2005 and 4/14/2006.

In a similar manner, the weekly composite chart can be constructed by concatenat-
ing the daily charts. In Figure 2.10, the average weekly activity between 1/1/2005 and 4/14/2006 is examined.

Composite charts are designed to assist traders in scheduling their primary trading
sessions. In accordance with our time definitions described at the beginning of this
chapter, the time frame in this chart is Sunday 00:00 through Friday 23:59, or six days. The time interval is measured in one- and two-hour increments.