R: timeSeries Class and Methods

TimeSeriesClass {fCalendar}

R Documentation

timeSeries Class and Methods

Description

A collection and description of functions and methods dealing with regular and irregular 'timeSeries' objects. Dates and times are implemented as 'timeDate' objects. Included are functions and methods for the generation and representation of 'timeSeries' objects, and for mathematical operations.

The functions and methods for the Generation of 'timeSeries' Objects are:

`timeSeries`	Creates a 'timeSeries' object from scratch,
`read.timeSeries`	Reads a 'timeSeries' from a spreadsheet file,
`as.timeSeries`	S3: Creates 'time Series' from a 'matrix',
`is.timeSeries`	S3: Tests if an object is of class a 'timeSeries',
`print.timeSeries`	S3: Print method for a 'timeSeries' object,
`plot.timeSeries`	S3: Plot method for a 'timeSeries' object,
`lines.timeSeries`	S3: Lines method for a 'timeSeries' object,
`Ops.timeSeries`	S3: Arith method for a 'timeSeries' object,
`[.timeSeries`	S3: "[" method for a 'timeSeries' object,
`head.timeSeries`	S3: returns the head of a 'timeSeries' object,
`tail.timeSeries`	S3: returns the tail of a 'timeSeries' object.

The functions and methods for the Representation of 'timeSeries' Objects are:

`seriesData`	Extracts data slot from a 'timeSeries',
`seriesPositions`	Extracts positions slot from a 'timeSeries',
`start.timeSeries`	S3: Extracts start date of a 'timeSeries',
`end.timeSeries`	S3: Extracts end date of a 'timeSeries',
`as.vector.timeSeries`	S3: Converts a 'timeSeries' to a vector,
`as.matrix.timeSeries`	S3: Converts a 'timeSeries' to a matrix,
`as.data.frame.timeSeries`	S3: Converts a 'timeSeries' to a data.frame.

The functions and methods for Math Operations of 'timeSeries' Objects are:

`applySeries`	Applies a function to margins of a 'timeSeries',
`alignDailySeries`	Aligns a daily 'timeSeries' to new positions,
`cutSeries`	Cuts out a piece from a 'timeSeries' object,
`mergeSeries`	Merges a 'timeSeries' object with a 'matrix',
`ohlcDailyPlot`	Plots open high low close bar chart,
`revSeries`	Reverts the order of 'timeSeries' object,
`diffSeries`	Takes differences from a 'timeSeries' object,
`lagSeries`	Lags a 'timeSeries' object,
`outlierSeries`	Removes outliers from a 'timeSeries' object,
`returnSeries`	Computes returns from a 'timeSeries' object,
`logSeries`	Returns logarithms of a 'timeSeries' object,
`absSeries`	Returns absolute values of a 'timeSeries' object.

Usage

timeSeries(data, charvec, units = NULL, format = "ISO", zone = "GMT", 
    FinCenter = myFinCenter, recordIDs = data.frame(), title = NULL, 
    documentation = NULL, ...) 
read.timeSeries(file, zone = "GMT", FinCenter = "", title = "", 
    documentation = "", sep = ";")
    
as.timeSeries(x, dimnames = TRUE, format = "")
is.timeSeries(object) 
## S3 method for class 'timeSeries':
print(x, ...)
## S3 method for class 'timeSeries':
plot(x, reference.grid = TRUE, lty = 1, ...) 
## S3 method for class 'timeSeries':
lines(x, ...)
## S3 method for class 'timeSeries':
Ops(e1, e2)
## S3 method for class 'timeSeries':
x[i = min(1, nrow(x@Data]):nrow(x@Data), 
    j = min(1, ncol(x@Data)):ncol(x@Data))
## S3 method for class 'timeSeries':
head(x, ...)
## S3 method for class 'timeSeries':
tail(x, ...)

seriesData(object)
seriesPositions(object)
## S3 method for class 'timeSeries':
start(x, ...)
## S3 method for class 'timeSeries':
end(x, ...)
## S3 method for class 'timeSeries':
as.vector(x, mode = "any")
## S3 method for class 'timeSeries':
as.matrix(x)
## S3 method for class 'timeSeries':
as.data.frame(x, row.names = NULL, optional = NULL)

applySeries(x, from = NULL, to = NULL, FUN = colAvgs, units = NULL, ...) 
cutSeries(x, from, to)
diffSeries(x, lag = 1, diff = 1, trim = FALSE, pad = NA)
lagSeries(x, k = 1, trim = FALSE, units = NULL)
outlierSeries(x, sd = 10, complement = TRUE)
mergeSeries(x, y, units = NULL)
returnSeries(x, type = c("continuous", "discrete"), percentage = FALSE, 
    trim = TRUE, digits = 4, units = NULL)
revSeries(x)
logSeries(x)
absSeries(x)
    
alignDailySeries(x, method = c("before", "after", "interp", "fillNA"), 
    include.weekends = FALSE, units = NULL)
ohlcDailyPlot(x, volume = TRUE, colOrder = c(1:5), units = 1e6, 
    xlab = c("Date", "Date"), ylab = c("Price", "Volume"), 
    main = c("O-H-L-C", "Volume"), grid.nx = 7, grid.lty = "solid", ...)

Arguments

`charvec`	a character vector of dates and times.
`complement`	[outlierSeries] - a logical flag, should the outler series or its complement be returns, by default `TRUE` which returns the series free of outliers.
`colOrder`	[ohlcDailyPlot] - an integer vector which gives the order of the prices and the volume in the input object. By default the following order of columns from 1 to 5 is assumed: Open, high, low, close, and volume.
`data`	a `data.frame` or a `matrix` object of numeric data.
`diff`	[diffSeries] - an integer indicating the order of the difference. By default 1.
`digits`	[returnSeries] - an integer value. The number of digits to be printed in the output.
`dimnames`	[as.timeSeries] - a logical, if `TRUE` the dimension names of the matrix are assigned to the time series object.
`documentation`	optional documentation string, or a vector of character strings.
`e1, e2`	[Ops] - two objects of class `timeSeries`.
`file`	the filename of a spreadsheet data set from which to import the data records.
`FinCenter`	a character with the the location of the financial center named as "continent/city".
`format`	the format specification of the input character vector, [as.timeSeries] - a character string with the format in POSIX notation to be passed to the time series object.
`from, to`	starting date and end date, `to` must be after `from`.
`FUN`	[applySeries] - a function to use for aggregation, by default `colAvgs`.
`grid.lty, grid.nx`	[ohlcDailyPlot] - The type of grid line and the number of grid lines used in the plot.
`i, j`	["["] - index arguments used for subsettings.
`include.weekends`	[alignDailySeries] - a logical value. Should weekend dates be included or removed from the series.
`k`	[lagSeries] - an integer value. The number of lags (in units of observations). By default 1.
`lag`	[diffSeries] - an integer indicating which lag to use. By default 1.
`lty`	[plot] - an integer value, the typ of line to be drawn. By default 1, a solid line.
`main`	[ohlcDailyPlot] - a character string to title the price and volume plot.
`method`	[alignDailySeries] - the method to be used for the alignment. A character string, one of `"before"`, use the data from the row whose position is just before the unmatched position, or `"after"`, use the data from the row whose position is just after the unmatched position, or `"linear"`, interpolate linearly between `"before"` and `"after"`.
`mode`	[as.vector] - A character string giving an atomic mode, or `"any"`.
`optional, row.names`	[as.data.frame] - both arguments are currently not used.
`object`	[is][seriesData][seriesPositions] - an object of class `timeSeries`.
`pad`	[diffSeries] - which value should get the padded values? By default `NA`. Another choice often used would be zero.
`percentage`	[returnSeries] - a logical value. By default `FALSE`, if `TRUE` the series will be expressed in percentage changes.
`recordIDs`	a data frame which can be used for record identification.
`reference.grid`	[plot] - a logical value. Should a grid be added to the plot? By default `TRUE`.
`sd`	[outlierSeries] - a numeric value of standard deviations, e.g. 10 means that values larger or smaller tahn ten times the standard deviation will be removed from the series.
`sep`	[read.timeSeries] - the field seperator used in the spreadsheet file to separate columns.
`title`	an optional title string, if not specified the inputs data name is deparsed.
`trim`	[diffSeries][returnSeries] - a logical value. By default `TRUE`, the first missing observation in the return series will be removed.
`type`	[returnSeries] - a character string specifying how to compute the returns. Valid choices are: `continuous` and `discrete`. For the default `type="continuous"`, the returns are calculated as the logarithmic differences, otherwise if `type="discrete"`, the returns are calculated as percentage changes.
`units`	[applySeries][lag][allignDailySeries][returnSeries][mergeSeries] - an optional character string, which allows to overwrite the current column names of a `timeSeries` object. By default `NULL` which means that the column names are selected automatically. [ohlcDailyPlot] - a numeric value, specifying in which multiples the volume should be referenced on the plot labels. By default 1e6, i.e. in units of 1 Million.
`volume`	[ohlcDailyPlot] - a logigical value. Should a volume plot added to the OHLC Plot. By default `TRUE`.
`x`	[as] - a `matrix` type object to be converted. [as.vector][as.matrix][as.data.frame] - [applySeries] - [cut][end][mergeSeries][plot][print][rev][start] - an object of class `timeSeries`.
`xlab, ylab`	[ohlcDailyPlot] - two string vectors to name the x and y axis of the price and volume plot.
`y`	[mergeSeries] - a `matrix` type object to be merged with an object of class `timeSeries`. Must have the same number of rows.
`zone`	the time zone or financial center where the data were recorded.
`...`	arguments passed to other methods.

Details

Generation of Time Series Objects:

We have defined a timeSeries class which is in many aspects similar to the S-Plus class with the same name, but has also some important differences. The class has seven Slots, the 'Data' slot which holds the time series data in matrix form, the 'position' slot which holds the time/date as a character vector, the 'format' and 'FinCenter' slots which are the same as for the 'timeDate' object, the 'units' slot which holds the column names of the data matrix, and a 'title' and a 'documentation' slot which hold descriptive character strings. Date and time is managed in the same way as for timeDate objects.

Value

timeSeries
read.timeSeries
as.timeSeries
return a S4 object of class timeSeries.

seriesData
seriesPositions
extract the @Data and @position slots from a timeSeries object. Thus, seriesData returns an object of class matrix, and seriesPositions returns an object of class timeDate.

is.timeSeries
returns TRUE or FALSE depending on whether its argument is of timeSeries type or not.

aggregateSeries
applySeries
cutSeries
mergeSeries
returnSeries
revSeries
return a S4 object of class timeSeries.

end, start
return a S4 object of class timedate. These are the start and end dates of a timeSeries object.

as.vector
as.matrix
as.data.frame
these are methods which convert a S4 object of class timeSeries either to a vector, a matrix or to a data frame.

plot
print
plot and print methods for an object of class timeSeries. Note that the plot function requires the packages its and Hmisc.

Note

These functions were written for Rmetrics users using R and Rmetrics under Microsoft's Windows operating system where timze zones, daylight saving times and holiday calendars are insuffeciently supported. The functions are untested for other system environments, but may be used.

Please, also be aware of the notes remarked in the help page describing the timeDate class.

IT IS IMPORTANT THAT YOU SET YOUR ENVIRONMENT VARIABLE TZ TO 'GMT', OTHERWISE SOME OF THE 'TIMEDATE' FUNCTIONS WOULD NOT PROPERLY WORK!

Author(s)

Diethelm Wuertz for the Rmetrics R-port.

Examples

## SOURCE("fCalendar.24A-TimeSeriesClass")

## data - 
   xmpBasics("\nStart: Load Microsoft Data > ")
   data(msft.dat)
   myFinCenter = "GMT"
   msft.dat[1:3, ]

## timeSeries -
   xmpBasics("\nNext: Create a timeSeries Object > ")
   close = msft.dat[, 5]
   ts = timeSeries(close, as.character(msft.dat[, 1]), format = "%m/%d/%Y", 
     units = "MSFT", zone = "GMT", FinCenter = "GMT")
   class(ts)
   ts[1:3,]
   c(start(ts), end(ts))
    
## cutSeries -    
   xmpBasics("\nNext: Cut out April Data from 2001 > ")
   ts.Apr01 = cutSeries(ts, "2001-04-01", "2001-04-30") 
   ts.Apr01

## returnSeries -  
   xmpBasics("\nNext: Compute Returns > ")
   # Compute Returns:
   args(returnSeries)
   # Continuous Returns:
   returnSeries(ts.Apr01)
   # Discrete Returns:
   returnSeries(ts.Apr01, type = "discrete")
   # Don't trim:
   returnSeries(ts.Apr01, trim = FALSE)
   # Use Percentage Values:
   returnSeries(ts.Apr01, percentage = TRUE, trim = FALSE)

## mergeSeries -    
   xmpBasics("\nNext: Merge Series with Returns > ")
   # Include last Day from March:
   ts.APR01 = cutSeries(ts, "2001-03-29", "2001-04-30") 
   ts.merged = mergeSeries(x = ts.APR01, 
     returnSeries(ts.APR01, trim = FALSE)@Data, units = c("MSFT", "Returns"))
   ts.merged
    
## alignDailySeries -     
   xmpBasics("\nNext: Align with NA > ")
   args(alignDailySeries)
   ts.ret = returnSeries(ts.APR01, trim = TRUE)
   GoodFriday(2001)   
   EasterMonday(2001) 
   alignDailySeries(ts.ret, method = "fillNA", include.weekends = FALSE)
   alignDailySeries(ts.ret, method = "fillNA", include.weekends = TRUE)
    
## alignDailySeries - 
   xmpBasics("\nNext: Interpolate > ")
   ts.ret
   alignDailySeries(ts.ret, method = "interp", include.weekend = FALSE)
   alignDailySeries(ts.ret, method = "interp", include.weekend = TRUE)

## applySeries -   
   xmpBasics("\nNext: Aggregate weekly > ")
   GoodFriday(2001)
   to = timeSequence(from = "2001-04-11", length.out = 3, 
     by = "weeks", format = "%Y-%m-%d", FinCenter = "GMT") 
   from = to - 6*24*3600
   data.frame(from, to)
   applySeries(ts.ret, from, to, FUN = sum)

[Package fCalendar version 221.10065 Index]