Three Black Crows 99

External links
• Stock Charts - Glossary (http://stockcharts.com/school/doku.php?id=chart_school:glossary_t)
• Investopedia - Dictionary (http://www.investopedia.com/terms/t/three_black_crows.asp)
• Prosticks - Chart Patterns (http://www.prosticks.com/education/candlestick_patterns/three_black_crows.php)

Morning star (candlestick pattern)

The Morning Star is a pattern seen in a
candlestick chart, a type of chart used by
stock analysts to describe and predict price
movements of a security, derivative, or
currency over time.

Description
The pattern is made up of three candles:
normally a long bearish candle, followed by
a short bullish or bearish doji, which is then
Illustration of the morningstar pattern
followed by a long bullish candle. In order
to have a valid Morning Star formation,
most traders will look for the top of the third candle to be at least half way up the body of the first candle in the
pattern. Black candles indicate falling prices, and white candles indicate rising prices.

Interpretation
When found in a downtrend, this pattern can be an indication that a reversal in the price trend is going to take place.
What the pattern represents from a supply and demand point of view is a lot of selling in the period which forms the
first black candle; then, a period of lower trading but with a reduced range, which indicates indecision in the market;
this forms the second candle. This is followed by a large white candle, representing buyers taking control of the
market. As the Morning Star is a three-candle pattern, traders oftentimes will not wait for confirmation from a fourth
candle before buying the stock. High volumes on the third trading day confirm the pattern. Traders will look at the
size of the candles for an indication of the size of the potential reversal. The larger the white and black candle, and
the higher the white candle moves in relation to the black candle, the larger the potential reversal.
The chart below illustrates.
Morning star (candlestick pattern) 100

The Morning Star pattern is circled. Note the high trading volumes on the third day.

The opposite occurring at the top of an uptrend is called an evening star.

Hikkake Pattern
The Hikkake Pattern (also known simply as Hikkake or an inside day false breakout), is a technical analysis
pattern used for determining market turning-points and continuations. It is a simple pattern that can be observed in
market price data, using traditional bar charts, or Japanese candlestick charts.

Description
The pattern consists of a measurable period of rest and volatility contraction in the market, followed by a relatively
brief price move that encourages unsuspecting traders and investors to adopt a false assumption regarding the likely
future direction of price. The pattern, once formed, yields its own set of trading parameters for the time and price of
market entry, the dollar risk amount (i.e., where to place protective stops), and the expected profit target. The pattern
is not meant as a stand alone "system" for market speculation, but rather as an ancillary technique to traditional
technical and fundamental market analysis methods.
The pattern is recognised in two variants, one bearish and one bullish. In both variants, the first bar of the pattern is
an inside bar (i.e., one which has both a higher low and a lower high, compared with the previous bar). This is then
followed by either a bar with both higher low and higher high for the bearish variant, or with lower low and lower
high for the bullish variant. Before the pattern produces a trading signal it must be confirmed; this happens when the
price passes below the low of the first bar of the pattern (in the bearish variant) or above the high of the first bar (in
the bullish variant). Confirmation must occur within three periods of the last bar of the signal for the signal to be
considered valid.
Hikkake Pattern 101

Origin
The hikkake pattern was first discovered and introduced to the financial community through a series of published
articles written by technical analyst.[1] The phrase "Hikkake" is a Japanese verb which means to "trick" or "ensnare."

Notable uses
Due to its popularity among institutional traders, the hikkake pattern has been adopted for use by INSTREAM,[2] the
Nordic power trading software company, in their E2 Energy Market Analysis Platform.
Growing interest in the hikkake pattern among traders and investors has also spawned attention from various book
authors. See: "Technical Analysis: The Complete Resource for Financial Market Technicians" by Charles D.
Kirkpatrick and Julie R. Dahlquist, and "Long/Short Market Dynamics: Trading Strategies for Today's Markets" by
Clive M. Corcoran.

References
[1] Daniel L. Chesler, CMT (http:/ / www. chesler. us)
[2] INTSTREAM (http:/ / www. intstream. com/ )

External links
• Trading False Moves with the Hikkake Pattern - PDF file (http://www.chesler.us/resources/articles/
chesler0404.pdf)
• Quantifying Market Deception with The Hikkake Pattern - PDF file (http://www.chesler.us/resources/articles/
TechAnalyst_Nov04.pdf)
• Noted technical analysis authority Thomas Bulkowski considers in detail the historical performance record of the
Hikkake pattern (http://thepatternsite.com/HikkakeBull.html)
 102

INDICATORS: Trend

Average Directional Index

The Average Directional Index (ADX) was developed in 1978 by J. Welles Wilder as an indicator of trend strength
in a series of prices of a financial instrument. ADX has become a widely used indicator for technical analysts, and is
provided as a standard in collections of indicators offered by various trading platforms.

Calculation
The ADX is a combination of two other indicators developed by Wilder, the positive directional indicator
(abbreviated +DI) and negative directional indicator (-DI). The ADX combines them and smooths the result with an
exponential moving average.
To calculate +DI and -DI, one needs price data consisting of high, low, and closing prices each period (typically each
day). One first calculates the Directional Movement (+DM and -DM):
UpMove = Today's High − Yesterday's High
DownMove = Yesterday's Low − Today's Low
if UpMove > DownMove and UpMove > 0, then +DM = UpMove, else +DM = 0
if DownMove > UpMove and DownMove > 0, then -DM = DownMove, else -DM = 0
After selecting the number of periods (Wilder used 14 days originally), +DI and -DI are:
+DI = 100 times exponential moving average of +DM divided by Average True Range
-DI = 100 times exponential moving average of -DM divided by Average True Range
The exponential moving average is calculated over the number of periods selected, and the average true range is an
exponential average of the true ranges. Then:
ADX = 100 times the exponential moving average of the Absolute value of (+DI − -DI) divided by (+DI +
-DI)
Variations of this calculation typically involve using different types of moving averages, such as a weighted moving
average or an adaptive moving average.

Interpretation
The ADX does not indicate trend direction, only trend strength. It is a lagging indicator; that is, a trend must have
established itself before the ADX will generate a signal that a trend is underway. ADX will range between 0 and
100. Generally, ADX readings below 20 indicate trend weakness, and readings above 40 indicate trend strength. An
extremely strong trend is indicated by readings above 50.

References
• J. Welles Wilder, Jr. (June 1978). New Concepts in Technical Trading Systems. Greensboro, NC: Trend Research.
ISBN 978-0894590276.
Ichimoku Kinkō Hyō 103

Ichimoku Kinkō Hyō

Ichimoku Kinko Hyo (一目均衡表 Ichimoku Kinkō Hyō) usually just called ichimoku is a technical analysis
method that builds on candlestick charting to improve the accuracy of forecasted price moves. It was developed in
the late 1930s by Goichi Hosoda (細田悟一 Hosoda Goichi), a Japanese journalist who used to be known as
Ichimoku Sanjin, which can be translated as "What a man in the mountain sees". He spent thirty years perfecting the
technique before releasing his findings to the general public in the late 1960s.[1]
Ichimoku Kinko Hyo translates to 'one glance equilibrium chart' or 'instant look at the balance chart' and is
sometimes referred to as 'one glance cloud chart' based on the unique 'clouds' that feature in ichimoku charting.[2]
Ichimoku is a moving average-based trend identification system and because it contains more data points than
standard candlestick charts, provides a clearer picture of potential price action.[3] The main difference between how
moving averages are plotted in ichimoku as opposed to other methods is that ichimoku's lines are constructed using
the 50% point of the highs and lows as opposed to the candle's closing price.
Ichimoku factors in time as an additional element along with the price action, similar to William Delbert Gann's
trading ideas.
Popular in Japan, ichimoku is gaining traction in the west through proponents of its charting accuracy such as
Lincoln FX and Ichi360.

The key elements of the ichimoku chart

Tenkan-sen
Tenkan-sen (転換線) calculation: (highest high + lowest low)/2 for the last 9 periods.
It is primarily used as a signal line and a minor support/resistance line.

Kijun-sen
Kijun-sen (基準線) calculation: (highest high + lowest low)/2 for the past 26 periods.
This is a confirmation line, a support/resistance line, and can be used as a trailing stop line.

Senkou span A
Senkou (先行) span A calculation: (Tenkan-sen + kijun-sen)/2 plotted 26 periods ahead.
Also called leading span 1, this line forms one edge of the kumo, or cloud

Senkou span B
Senkou span B calculation: (highest high + lowest low)/2 calculated over the past 52 time periods and plotted 26
periods ahead.
Also called leading span 2, this line forms the other edge of the kumo.

Kumo
Kumo (雲, cloud) is the space between senkou span A and B. The cloud edges identify current and potential future
support and resistance points.
Ichimoku Kinkō Hyō 104

Chikou span
Chikou (遅行) span calculation: today's closing price projected back 26 days on the chart.
Also called the lagging span it is used as a support/resistance aid.

References
[1] Kumotrader Ichimoku Wiki (http:/ / www. kumotrader. com/ ichimoku_wiki/ index. php?title=Main_Page)
[2] Example ichimoku chart showing all the elements (http:/ / www. lincolnfx. com/ forex-forecasts/ 2010/ 01/ 24/ ichimoku-elements-2/ )
[3] Ichimoku Cloud Filters Information Storm (http:/ / www. investopedia. com/ articles/ forex/ 06/ ichimoku. asp)

MACD
MACD (Moving Average Convergence/Divergence) is a technical analysis indicator created by Gerald Appel in the
late 1970s.[1] It is used to spot changes in the strength, direction, momentum, and duration of a trend in a stock's
price.
The MACD is a computation of the difference between two exponential moving averages (EMAs) of closing prices.
This difference is charted over time, alongside a moving average of the difference. The divergence between the two
is shown as a histogram or bar graph.
Exponential moving averages highlight recent changes in a stock's price. By comparing EMAs of different periods,
the MACD line illustrates changes in the trend of a stock. Then by comparing that difference to an average, an
analyst can chart subtle shifts in the stock's trend.
Since the MACD is based on moving averages, it is inherently a lagging indicator. As a metric of price trends, the
MACD is less useful for stocks that are not trending or are trading erratically.
Note that the term "MACD" is used both generally, to refer to the indicator as a whole, and specifically, to the
MACD line itself.

Basic components

MACD 12,26,9

The graph above shows a stock with a MACD indicator underneath it. The indicator shows a blue line, a red line,
and a histogram or bar chart which calculates the difference between the two lines. Values are calculated from the
price of the stock in the main part of the graph.
For the example above this means:
• MACD line (blue line): difference between the 12 and 26 days EMAs
• signal (red line): 9 day EMA of the blue line
• histogram (bar graph): difference between the blue and red lines
Mathematically:
MACD 105

• MACD = EMA[fast,12] – EMA[slow, 26]

• signal = EMA[period,9] of MACD
• histogram = MACD – signal
The period for the moving averages on which an MACD is based can vary, but the most commonly used parameters
involve a faster EMA of 12 days, a slower EMA of 26 days, and the signal line as a 9 day EMA of the difference
between the two. It is written in the form, MACD(faster, slower, signal) or in this case, MACD(12,26,9).

Interpretation
Exponential moving averages highlight recent changes in a stock's price. By comparing EMAs of different lengths,
the MACD line gauges changes in the trend of a stock. By then comparing differences in the change of that line to an
average, an analyst can identify subtle shifts in the strength and direction of a stock's trend.
Traders recognize three meaningful signals generated by the MACD indicator.
When:
• the MACD line crosses the signal line
• the MACD line crosses zero
• there is a divergence between the MACD line and the price of the stock or between the histogram and the price of
the stock
Graphically this corresponds to:
• the blue line crossing the red line
• the blue line crossing the x-axis (the straight black line in the middle of the indicator)
• higher highs (lower lows) on the price graph but not on the blue line, or higher highs (lower lows) on the price
graph but not on the bar graph
And mathematically:
• MACD – signal = 0
• EMA[fast,12] – EMA[slow,26] = 0
• Sign (relative price extremumfinal – relative price extremuminitial) ≠ Sign (relative MACD extremumfinal –
MACD extremuminitial)

Signal line crossover

Signal line crossovers are the primary cues provided by the MACD. The standard interpretation is to buy when the
MACD line crosses up through the signal line, or sell when it crosses down through the signal line.
The upwards move is called a bullish crossover and the downwards move a bearish crossover. Respectively, they
indicate that the trend in the stock is about to accelerate in the direction of the crossover.
The histogram shows when a crossing occurs. Since the histogram is the difference between the MACD line and the
signal line, when they cross there is no difference between them.
The histogram can also help in visualizing when the two lines are approaching a crossover. Though it may show a
difference, the changing size of the difference can indicate the acceleration of a trend. A narrowing histogram
suggests a crossover may be approaching, and a widening histogram suggests that an ongoing trend is likely to get
even stronger.
While it is theoretically possible for a trend to increase indefinitely, under normal circumstances, even stocks
moving drastically will eventually slow down, lest they go up to infinity or down to nothing.
MACD 106

Zero crossover
A crossing of the MACD line through zero happens when there is no difference between the fast and slow EMAs. A
move from positive to negative is bearish and from negative to positive, bullish. Zero crossovers provide evidence of
a change in the direction of a trend but less confirmation of its momentum than a signal line crossover.

Divergence
The third cue, divergence, refers to a discrepancy between the MACD line and the graph of the stock price. Positive
divergence between the MACD and price arises when price hits a new low, but the MACD doesn't. This is
interpreted as bullish, suggesting the downtrend may be nearly over. Negative divergence is when the stock price hits
a new high but the MACD does not. This is interpreted as bearish, suggesting that recent price increases will not
continue.
Divergence may also occur between the stock price and the histogram. If new high price levels are not confirmed by
new high histogram levels, it is considered bearish; alternately, if new low price levels are not confirmed by new low
histogram levels, it is considered bullish.
Longer and sharper divergences—distinct peaks or troughs—are regarded as more significant than small, shallow
patterns.

Timing
The MACD is only as useful as the context in which it is applied. An analyst might apply the MACD to a weekly
scale before looking at a daily scale, in order to avoid making short term trades against the direction of the
intermediate trend.[2] Analysts will also vary the parameters of the MACD to track trends of varying duration. One
popular short-term set-up, for example, is the (5,35,5).

False signals
Like any indicator, the MACD can generate false signals. A false positive, for example, would be a bullish crossover
followed by a sudden decline in a stock. A false negative would be a situation where there was no bullish crossover,
yet the stock accelerated suddenly upwards.
A prudent strategy would be to apply a filter to signal line crossovers to ensure that they will hold. An example of a
price filter would be to buy if the MACD line breaks above the signal line and then remains above it for three days.
As with any filtering strategy, this reduces both the probability of false signals as well as the frequency of missed
profit.
Analysts use a variety of approaches to filter out false signals and confirm true ones. As a lagging indicator, the
MACD is often paired with a leading indicator, like the Relative Strength Index (RSI). Historical comparisons to
similar stocks as well as a careful investigation of past price movements provide added information about how a
stock tends to move.
MACD 107

Limitations
The MACD has often been criticized for failing to respond in very low or alternately very high volatility market
conditions.[3] Since the MACD measures the divergence between averages, it can give meaningful feedback only as
trends change. Thus, the MACD is less useful if the market is not trending, that is, if it is trading sideways, or if the
market is trading erratically, making sudden, dramatic, or countervailing moves.
In a sideways market, the divergence between averages will not have a trend to illuminate. In an erratic market, the
changes will happen too quickly to be picked up by moving averages or will cancel each other out, diminishing the
MACDs usefulness. A partial caveat to this criticism is that whether a market is trending or volatile is always
relative to a particular timeframe, and the MACD can be adjusted to shorter or longer spans.
Finally, though some analysts trade on technical indicators alone, the abundance of experts recommend a complete
work-up of a company's business sectors, financial strength, past earnings, new products, management, and
institutional buying. For more traditional investors, an indicator like the MACD may be used only to support a
previously determined stock choice, or to select an ideal entry-point into a fundamentally sound stock.

Oscillator classification
The MACD is an absolute price oscillator (APO), because it deals with the actual prices of moving averages rather
than percentage changes. A percentage price oscillator (PPO), on the other hand, computes the difference between
two moving averages of price divided by the longer moving average value.
While an APO will show greater levels for higher priced securities and smaller levels for lower priced securities, a
PPO calculates changes relative to price. Subsequently, a PPO is preferred when: comparing oscillator values
between different securities, especially those with substantially different prices; or comparing oscillator values for
the same security at significantly different times, especially a security whose value has changed greatly.
A third member of the price oscillator family is the detrended price oscillator (DPO), which ignores long term trends
while emphasizing short term patterns.

Signal processing theory

In signal processing terms, the MACD is a filtered measure of velocity. The velocity has been passed through two
first-order linear low pass filters. The "signal line" is that resulting velocity, filtered again. The difference between
those two, the histogram, is a measure of the acceleration, with all three filters applied. An MACD crossover of the
signal line indicates that the direction of the acceleration is changing. The MACD line crossing zero suggests that the
average velocity is changing direction.

History
The MACD was invented by Gerald Appel in the 1970's. Thomas Aspray added a histogram to the MACD in 1986,
as a means to anticipate MACD crossovers, an indicator of important moves in the underlying security.

References
[1] Appel, Gerald (1999). Technical Analysis Power Tools for Active Investors. Financial Times Prentice Hall. pp. 166. ISBN 0131479024.
[2] Murphy, John (1999). Technical Analysis of the Financial Markets. Prentice Hall Press. pp. 252–255. ISBN 0735200661.
[3] Fidelity Trend Trading with Short-Term Patterns (http:/ / personal. fidelity. com/ myfidelity/ atn/ archives/ august2003. html) Retrieved on
May 12, 2007
MACD 108

External links
• Interpreting MACD Crossovers, Histograms, and Divergences (http://www.onlinetradingconcepts.com/
TechnicalAnalysis/MACD.html)
• MACD page (http://www.investopedia.com/terms/m/macd.asp), at Investopedia.com
• MACD page (http://stockcharts.com/school/doku.
php?id=chart_school:technical_indicators:moving_average_conve), at Stockcharts.com

Mass index
The mass index is an indicator, developed by Donald Dorsey, used in technical analysis to predict trend reversals. It
is based on the notion that there is a tendency for reversal when the price range widens, and therefore compares
previous trading ranges (highs minus lows).
Mass index for a commodity is obtained[1] by calculating its exponential moving average over a 9 day period and the
exponential moving average of this average (a "double" average), and summing the ratio of these two over a given
amount of days (usually 25).

Generally the EMA and the re-smoothed EMA of EMA are fairly close, making their ratio is roughly 1 and the sum
around 25.
According to Dorsey, a so-called "reversal bulge" is a probable signal of trend reversal (regardless of the trend's
direction).[2] Such a bulge takes place when a 25-day mass index reaches 27.0 and then falls to below 26 (or 26.5). A
9-day prime moving average is usually used to determine whether the bulge is a buy or sell signal.
This formula uses intraday range values: not the "true range," which adjusts for full and partial gaps. Also, the
"bulge" does not indicate direction.

References
[1] Mass Index construction (http:/ / www. incrediblecharts. com/ technical/ mi_construction. htm) at IncredibleCharts.com
[2] Mass Index (http:/ / www. incrediblecharts. com/ technical/ mass_index. htm) at IncredibleCharts.com
Moving average 109

Moving average
In statistics, a moving average, also called rolling average, rolling mean or running average, is a type of finite
impulse response filter used to analyze a set of data points by creating a series of averages of different subsets of the
full data set.
Given a series of numbers and a fixed subset size, the moving average can be obtained by first taking the average of
the first subset. The fixed subset size is then shifted forward, creating a new subset of numbers, which is averaged.
This process is repeated over the entire data series. The plot line connecting all the (fixed) averages is the moving
average. Thus, a moving average is not a single number, but it is a set of numbers, each of which is the average of
the corresponding subset of a larger set of data points. A moving average may also use unequal weights for each data
value in the subset to emphasize particular values in the subset.
A moving average is commonly used with time series data to smooth out short-term fluctuations and highlight
longer-term trends or cycles. The threshold between short-term and long-term depends on the application, and the
parameters of the moving average will be set accordingly. For example, it is often used in technical analysis of
financial data, like stock prices, returns or trading volumes. It is also used in economics to examine gross domestic
product, employment or other macroeconomic time series. Mathematically, a moving average is a type of
convolution and so it is also similar to the low-pass filter used in signal processing. When used with non-time series
data, a moving average simply acts as a generic smoothing operation without any specific connection to time,
although typically some kind of ordering is implied.

Simple moving average

A simple moving average (SMA) is the unweighted mean of the previous n data points. For example, a 10-day
simple moving average of closing price is the mean of the previous 10 days' closing prices. If those prices are
then the formula is

When calculating successive values, a new value comes into the sum and an old value drops out, meaning a full
summation each time is unnecessary,

In technical analysis there are various popular values for n, like 10 days, 40 days, or 200 days. The period selected
depends on the kind of movement one is concentrating on, such as short, intermediate, or long term. In any case
moving average levels are interpreted as support in a rising market, or resistance in a falling market.
In all cases a moving average lags behind the latest data point, simply from the nature of its smoothing. An SMA can
lag to an undesirable extent, and can be disproportionately influenced by old data points dropping out of the average.
This is addressed by giving extra weight to more recent data points, as in the weighted and exponential moving
averages.
One characteristic of the SMA is that if the data have a periodic fluctuation, then applying an SMA of that period
will eliminate that variation (the average always containing one complete cycle). But a perfectly regular cycle is
rarely encountered in economics or finance.[1]
For a number of applications it is advantageous to avoid the shifting induced by using only 'past' data. Hence a
central moving average can be computed, using both 'past' and 'future' data. The 'future' data in this case are not
predictions, but merely data obtained after the time at which the average is to be computed.
Moving average 110

Cumulative moving average

The cumulative moving average is also frequently called a running average or a long running average although the
term running average is also used as synonym for a moving average. This article uses the term cumulative moving
average or simply cumulative average since this term is more descriptive and unambiguous.
In some data acquisition systems, the data arrives in an ordered data stream and the statistician would like to get the
average of all of the data up until the current data point. For example, an investor may want the average price of all
of the stock transactions for a particular stock up until the current time. As each new transaction occurs, the average
price at the time of the transaction can be calculated for all of the transactions up to that point using the cumulative
average. This is the cumulative average, which is typically an unweighted average of the sequence of i values x1, ...,
xi up to the current time:

The brute force method to calculate this would be to store all of the data and calculate the sum and divide by the
number of data points every time a new data point arrived. However, it is possible to simply update cumulative
average as a new value xi+1 becomes available, using the formula:

where CA0 can be taken to be equal to 0.

Thus the current cumulative average for a new data point is equal to the previous cumulative average plus the
difference between the latest data point and the previous average divided by the number of points received so far.
When all of the data points arrive (i = N), the cumulative average will equal the final average.
The derivation of the cumulative average formula is straightforward. Using

and similarly for i + 1, it is seen that

Solving this equation for CAi+1 results in:

Weighted moving average

A weighted average is any average that has multiplying factors to give different weights to different data points.
Mathematically, the moving average is the convolution of the data points with a moving average function; in
technical analysis, a weighted moving average (WMA) has the specific meaning of weights that decrease
arithmetically. In an n-day WMA the latest day has weight n, the second latest n − 1, etc, down to one.
Moving average 111

WMA weights n = 15

The denominator is a triangle number, and can be easily computed as

When calculating the WMA across successive values, it can be noted the difference between the numerators of
WMAM+1 and WMAM is npM+1 − pM − ... − pM−n+1. If we denote the sum pM + ... + pM−n+1 by TotalM, then

The graph at the right shows how the weights decrease, from highest weight for the most recent data points, down to
zero. It can be compared to the weights in the exponential moving average which follows.

Exponential moving average

An exponential moving average (EMA), also known as an
exponentially weighted moving average (EWMA),[2] is a type of
infinite impulse response filter that applies weighting factors which
decrease exponentially. The weighting for each older data point
decreases exponentially, never reaching zero. The graph at right shows
an example of the weight decrease.

The formula for calculating the EMA at time periods t > 2 is

EMA weights N=15

Where:
• The coefficient α represents the degree of weighting decrease, a constant smoothing factor between 0 and 1. A
higher α discounts older observations faster. Alternatively, α may be expressed in terms of N time periods, where
α = 2/(N+1). For example, N = 19 is equivalent to α = 0.1. The half-life of the weights (the interval over which the
weights decrease by a factor of two) is approximately N/2.8854 (within 1% if N > 5).
• Yt is the observation at a time period t.
• St' is the value of the EMA at any time period t.
S1 is undefined. S2 may be initialized in a number of different ways, most commonly by setting S2 to Y1, though
other techniques exist, such as setting S2 to an average of the first 4 or 5 observations. The prominence of the S2
initialization's effect on the resultant moving average depends on α; smaller α values make the choice of S2 relatively
Moving average 112

more important than larger α values, since a higher α discounts older observations faster.
This formulation is according to Hunter (1986)[3] . By repeated application of this formula for different times, we can
eventually write St as a weighted sum of the data points Yt, as:

for any suitable k = 0, 1, 2, ... The weight of the general data point is .
[4]
An alternate approach by Roberts (1959) uses Yt in lieu of Yt−1 :

This formula can also be expressed in technical analysis terms as follows, showing how the EMA steps towards the
latest data point, but only by a proportion of the difference (each time):[5]

Expanding out each time results in the following power series, showing how the weighting factor on
each data point p1, p2, etc, decreases exponentially:
[6]

This is an infinite sum with decreasing terms.

The N periods in an N-day EMA only specify the α factor. N is not a stopping point for the calculation in the way it
is in an SMA or WMA. For sufficiently large N, The first N data points in an EMA represent about 86% of the total
weight in the calculation[7] :

i.e. simplified[8] , tends to .

The power formula above gives a starting value for a particular day, after which the successive days formula shown
first can be applied. The question of how far back to go for an initial value depends, in the worst case, on the data. If
there are huge p price values in old data then they'll have an effect on the total even if their weighting is very small.
If one assumes prices don't vary too wildly then just the weighting can be considered. The weight omitted by
stopping after k terms is

which is

i.e. a fraction

out of the total weight.

For example, to have 99.9% of the weight, set above ratio equal to 0.1% and solve for k:

terms should be used. Since approaches as N increases[9] , this simplifies to approximately[10]

Moving average 113

for this example (99.9% weight).

Modified moving average

A modified moving average (MMA), running moving average (RMA), or smoothed moving average is defined
as:

In short, this is exponential moving average, with .

Application to measuring computer performance

Some computer performance metrics, e.g. the average process queue length, or the average CPU utilization, use a
form of exponential moving average.

Here is defined as a function of time between two readings. An example of a coefficient giving bigger weight to
the current reading, and smaller weight to the older readings is

where time for readings tn is expressed in seconds, and is the period of time in minutes over which the reading is
said to be averaged (the mean lifetime of each reading in the average). Given the above definition of , the moving
average can be expressed as

For example, a 15-minute average L of a process queue length Q, measured every 5 seconds (time difference is 5
seconds), is computed as

Other weightings
Other weighting systems are used occasionally – for example, in share trading a volume weighting will weight each
time period in proportion to its trading volume.
A further weighting, used by actuaries, is Spencer's 15-Point Moving Average[11] (a central moving average). The
symmetric weight coefficients are -3, -6, -5, 3, 21, 46, 67, 74, 67, 46, 21, 3, -5, -6, -3.

Moving median
From a statistical point of view, the moving average, when used to estimate the underlying trend in a time series, is
susceptible to rare events such as rapid shocks or other anomalies. A more robust estimate of the trend is the simple
moving median over n time points:

where the median is found by, for example, sorting the values inside the brackets and finding the value in the middle.
Statistically, the moving average is optimal for recovering the underlying trend of the time series when the
fluctuations about the trend are normally distributed. However, the normal distribution does not place high
probability on very large deviations from the trend which explains why such deviations will have a
disproportionately large effect on the trend estimate. It can be shown that if the fluctuations are instead assumed to
be Laplace distributed, then the moving median is statistically optimal[12] . For a given variance, the Laplace
distribution places higher probability on rare events than does the normal, which explains why the moving median
Moving average 114

tolerates shocks better than the moving mean.

When the simple moving median above is central, the smoothing is identical to the median filter which has
applications in, for example, image signal processing.

Notes and references

[1] Statistical Analysis, Ya-lun Chou, Holt International, 1975, ISBN 0030894220, section 17.9.
[2] http:/ / lorien. ncl. ac. uk/ ming/ filter/ filewma. htm
[3] NIST/SEMATECH e-Handbook of Statistical Methods: Single Exponential Smoothing (http:/ / www. itl. nist. gov/ div898/ handbook/ pmc/
section4/ pmc431. htm) at the National Institute of Standards and Technology
[4] NIST/SEMATECH e-Handbook of Statistical Methods: EWMA Control Charts (http:/ / www. itl. nist. gov/ div898/ handbook/ pmc/
section3/ pmc324. htm) at the National Institute of Standards and Technology
[5] Moving Averages page (http:/ / www. stockcharts. com/ education/ IndicatorAnalysis/ indic_movingAvg. html) at StockCharts.com

[6] ,

since .
[7] The denominator on the left-hand side should be unity, and the numerator will become the right-hand side (geometric series),

[8] Because (1+x/n)n becomes ex for large n.

[9] It means -> 0, and the Taylor series of tends to .
[10] loge(0.001) / 2 = -3.45
[11] Spencer's 15-Point Moving Average — from Wolfram MathWorld (http:/ / mathworld. wolfram. com/ Spencers15-PointMovingAverage.
html)
[12] G.R. Arce, "Nonlinear Signal Processing: A Statistical Approach", Wiley:New Jersey, USA, 2005.

External links
• EWMA in determining network traffic and ethernet (http://www.think-lamp.com/2009/03/
the-hidden-power-of-ping/)
• FastMedFilt1D: Fast Matlab software for computing the simple moving median of a time series. (http://www.
physics.ox.ac.uk/users/littlem/software/)
,
Parabolic SAR 115

Parabolic SAR
In the field of technical analysis,
Parabolic SAR (SAR - stop and
reverse) is a method devised by J.
Welles Wilder, Jr., to find trends in
market prices or securities. It may be
used as a trailing stop loss based on
prices tending to stay within a
parabolic curve during a strong trend.

The concept draws on the idea that

time is the enemy (similar to option
theory's concept of time decay), and
unless a security can continue to
generate more profits over time, it
should be liquidated. The indicator Parabolic SAR for Ebay during 2002.
generally works well in trending
markets, but provides "whipsaws" during non-trending, sideways phases; as such, Wilder recommended establishing
the strength and direction of the trend first through the use of things such as the Average Directional Index, and then
using the Parabolic SAR to trade that trend.

A parabola below the price is generally bullish, while a parabola above is generally bearish.

Construction
The Parabolic SAR is calculated almost independently for each trend in the price. When the price is in an uptrend,
the SAR appears below the price and converges upwards towards it. Similarly, on a downtrend, the SAR appears
above the price and converges downwards.
At each step within a trend, the SAR is calculated ahead of time. That is, tomorrow's SAR value is built using data
available today. The general formula used for this is:

Where and represent today's and tomorrow's SAR values, respectively.

The extreme point, , is a record kept during each trend that represents the highest value reached by the price
during the current uptrend — or lowest value during a downtrend. On each period, if a new maximum (or minimum)
is observed, the EP is updated with that value.
The value represents the acceleration factor. Usually, this is set to a value of 0.02 initially. This factor is
increased by 0.02 each time a new EP is recorded. In other words, each time a new EP is observed, it will increase
the acceleration factor. This will then quicken the rate at which the SAR converges towards the price. To keep it
from getting too large, a maximum value for the acceleration factor is normally set at 0.20, so that it never goes
beyond that. For stocks trading, it is preferable to set the acceleration factor to 0.01, in order to be less sensitive to
local decreases. For commodity or currency trading, it is preferable to use a value of 0.02.
The SAR is iteratively calculated for each new period using this recursive definition. There are, however, two special
cases that will modify the SAR value:
• If tomorrow's SAR value lies within (or beyond) today's or yesterday's price range, the SAR must be set to the
closest price bound. For example, if in an uptrend, the new SAR value is calculated and it results to be greater
than today's or yesterday's lowest price, the SAR must be set equal to that lower boundary.