Create A New TA Indicator For chartSeries
Functions to assist in the creation of indicators or content to be drawn on plots produced by chartSeries.
addTA(ta,
order = NULL,
on = NA,
legend = "auto",
yrange = NULL,
...)
newTA(FUN,
preFUN,
postFUN,
on = NA,
yrange = NULL,
legend.name,
fdots = TRUE,
cdots = TRUE,
data.at = 1,
...)ta |
data to be plotted |
order |
which should the columns (if > 1) be plotted |
legend |
what custom legend text should be added to the chart. |
FUN |
Main filter function name - as a symbol |
preFUN |
Pre-filter transformation or extraction function |
postFUN |
Post-filter transformation or extraction function |
on |
where to draw |
yrange |
length 2 vector of y-axis range |
legend.name |
optional legend heading, automatically derived otherwise |
fdots |
should any ... be included in the main filter call |
cdots |
should any ... be included in the resultant function object.
|
data.at |
which arguement to the main filter function is for data. |
... |
any additonal graphical parameters/default to be included. |
Both addTA and newTA can be used to
dynamically add custom content to a displayed chart.
addTA takes a series of values, either in
a form coercible to xts or of the same length as
the charted series has rows, and displays the results in
either a new TA sub-window, or over/underlayed on
the main price chart. If the object can be coerced to
xts, the time values present must only be within
the original series time-range. Internally a merge
of dates occurs and will allow for the plotting
of discontinuous series.
The order argument allows for multiple column
data to be plotted in an order that makes the most visual
sense.
Specifying a legend will override the standard
parsing of the addTA call to attempt a guess at a suitable
title for the sub-chart. Specifying this will cause the standard
last value to not be printed.
The ... arg to addTA is used to set graphical parameters
interpretable by lines.
newTA acts as more of a skeleton function, taking
functions as arguments, as well as charting parameters,
and returns a function that can be called in the same
manner as the built-in TA tools, such as addRSI and
addMACD. Essentially a dynamic code generator
that allows for highly customizable chart tools with
minimal (possibly zero) coding. It is also possible
to modify the resultant code to further change behavior.
To create a new TA function with newTA certain arguments
must be specified.
The FUN argument is a function symbol (or coercible to such)
that is the primary filter to be used on the core-data of a chartSeries
chart. This can be like most of the functions
within the TTR package — e.g. RSI or EMA. The resultant object
of the function call will be equal to calling the function
on the original data passed into the chartSeries function that created the
chart. It should be coercible to a matrix object, of one or more
columns of output. By default all columns of output will be added to the chart,
unless suppressed by passing the appropriately positioned type='n' as
the ... arg. Note that this will not suppress the labels added to the chart.
The preFUN argument will be called on the main chart's data prior
to passing it to FUN. This must be a function symbol or a character
string of the name function to be called.
The postFUN argument will be called on the resultant data
returned from the FUN filter. This is useful for extracting
the relevant data from the returned filter data. Like preFUN
it must be a function symbol or a character string of the name
of the function to be called.
The yrange argument is used to provide a custom
scale to the y-axis. If NULL the min and
max of the data to be plotted will be used for the y-axis
range.
The on is used to identify which subchart to add the
graphic to. By default, on=NA will draw the series
in a new subchart below the last indicator. Setting this
to either a positive or negative value will allow for
the series to be super-imposed on, or under, the (sub)chart
specified, respectively. A value of 1 refers to the main chart, and at present
is the only location supported.
legend.name will change the main label for a new plot.
fdots and cdots enable inclusion or suppression
of the ... within the resulting TA code's call to FUN,
or the argument list of the new TA function, respectively.
In order to facilitate user-specified graphical
parameters it is usually desireable to not impose
artificial limits on the end-user
with constraints on types of parameters available.
By default the new TA function will include the dots
argument, and the internal FUN call will keep all arguments, including
the dots. This may pose issues if the internal function then
passes those ... arguments to a function that can't handle them.
The final argument is data.at which is the position
in the FUN argument list which expects the data
to be passed in at. This default to the sensible
first position, though can be changed at the time of creation
by setting this argument to the required value.
While the above functions are usually sufficient
to construct very pleasing graphical additions to
a chart, it may be necessary to modify by-hand
the code produced. This can be accomplished by
dumping the function to a file, or using fix
on it during an interactive session.
Another item of note, with respect to newTA is the
naming of the main legend label. Following
addTA convention, the first ‘add’ is stripped
from the function name, and the rest of the call's
name is used as the label. This can be overridden
by specifying legend.name in the construction
of the new TA call, or by passing legend into
the new TA function. Subtle differences exist, with
the former being the preferred solution.
While both functions can be used to build new indicators without any understanding of the internal chartSeries process, it may be beneficial in more complex cases to have a knowledge of the multi-step process involved in creating a chart via chartSeries.
to be added...
addTA will invisibly return an S4 object of
class chobTA. If this function is called
interactively, the chobTA object will be
evaluated and added to the current chart.
newTA will return a function object that
can either be assigned or evaluated. Evaluating
this function will follow the logic of any standard
addTA-style call, returning invisibly a chobTA
object, or adding to the chart.
Both interfaces are meant to fascilitate custom
chart additions. addTA is for
adding any arbitrary series to a chart, where-as
newTA works with the underlying series with
the main chart object. The latter also
acts as a dynamic TA skeleton generation tool
to help develop reusable TA generation code
for use on any chart.
Jeffrey A. Ryan
chartSeries,
TA,
chob,
chobTA
## Not run:
getSymbols('SBUX')
barChart(SBUX)
addTA(EMA(Cl(SBUX)), on=1, col=6)
addTA(OpCl(SBUX), col=4, type='b', lwd=2)
# create new EMA TA function
newEMA <- newTA(EMA, Cl, on=1, col=7)
newEMA()
newEMA(on=NA, col=5)
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.