Companion to Tsay (2005) Analysis of Financial Time Series

Description

R companion to Tsay (2005) Analysis of Financial Time Series, second edition (Wiley). Includes data sets, functions and script files required to work some of the examples. Version 0.3-x includes R objects for all data files used in the text and script files to recreate most of the analyses in chapters 1-3 and 9 plus parts of chapters 4 and 11.

Details

Index:

ARIMA                   Arima with Ljung-Box
Acf                     Autocorrelation Function
ArchTest                ARCH LM Test
AutocorTest             Box-Ljung autocorrelation test
FinTS-package           Companion to Tsay (2005) Analysis of Financial
                        Time Series
FinTS.stats             Financial Time Series summary statistics
TsayFiles               List of the names of files downloaded from the
                        "Analysis of Financial Data" web site.
apca                    Asymptotic PCA
as.yearmon2             Conditionally convert x to yearmon if the
                        conversion is unique, retaining x as names.
ch01data                Financial time series for Tsay (2005, ch. 1)
ch02data                Financial time series for Tsay (2005, ch. 2)
ch03data                Financial time series for Tsay (2005, ch. 3)
ch04data                Financial time series for Tsay (2005, ch. 4)
ch05data                Financial time series for Tsay (2005, ch. 5)
ch06data                Financial time series for Tsay (2005, ch. 6)
ch07data                Financial time series for Tsay (2005, ch. 7)
ch08data                Financial time series for Tsay (2005, ch. 8)
ch09data                Financial time series for Tsay (2005, ch. 9)
ch10data                Financial time series for Tsay (2005, ch. 10)
ch11data                Financial time series for Tsay (2005, ch. 11)
ch12data                Financial time series for Tsay (2005, ch. 12)
compoundInterest        Compute compound interest
findConjugates          Find complex conjugate pairs
package.dir             Directory of a package
plot.loadings           Plot loadings
plotArmaTrueacf         Plot the theoretical ACF corresponding to an
                        ARMA model
read.yearmon            Reading Monthly zoo Series
runscript               Run a package script
url2data                Create local copies of files read from urls

See the scripts subdirectory of the FinTS installation directory = system.file(package='FinTS').

Corrections to the script files provided and contributions to script files for other chapters will be graciously accepted.

Author(s)

Spencer Graves

Maintainer: Georgi N. Boshnakov <georgi.boshnakov@manchester.ac.uk>

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley)

See Also

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

Examples

# Where is the 'FinTS' directory?
system.file(package='FinTS')

# View the script file 'ch01.R', which is in the 'scripts'
# subdirectory of the system.file(package='FinTS') directory:
runscript(1, 'view')

# SP statistics in Table 1.2 of Tsay
data(d.ibmvwewsp6203)
FinTS.stats(100*d.ibmvwewsp6203[, "SP"])

Arima with Ljung-Box

Description

Fit an ARIMA model and test residuals with the Ljung-Box statistic

Usage

ARIMA(x, order = c(0, 0, 0),
     seasonal = list(order = c(0, 0, 0), period = NA),
     xreg = NULL, include.mean = TRUE, transform.pars = TRUE,
     fixed = NULL, init = NULL, method = c("CSS-ML", "ML", "CSS"),
     n.cond, optim.control = list(), kappa = 1e6, Box.test.lag=NULL,
     Box.test.df = c("net.lag", "lag"),
     type = c("Ljung-Box", "Box-Pierce", "rank"))

Arguments

Details

1. Fit the desired model using 'arima'.

2. Compute the desired number of lags for Box.test

3. Apply 'AutocorTest' to the whitened residuals.

NOTE: Some software does not adjust the degrees of freedom for the number of parameters estimated. Tsay (2005) and Enders (2004) do. The need to adjust the degrees of freedom discussed by Brockwell and Davis (1990), who provide a proof describing the circumstances under which this is appropriate.

This is, however, an asymptotic result, and it would help to have simulation studies of the distribution of the Ljung-Box statistic, estimating degrees of freedom and evaluating goodness of fit. Burns recommends a rank version of the Ljung-Box test, but does not estimate degrees of freedom. If you have done such a simulation or know of a reference describing such, would you please notify the maintainer of this package?

4. If 'xreg' is supplied, compute r.squared.

Value

an 'arima' object with an additional 'Box.test' component and if 'xreg' is not null, an 'r.squared' component.

NOTE: The 'Box.test' help page in R 2.6.1 says, 'Missing values are not handled.' However, if 'x' contains NAs, 'ARIMA' still returns a numeric answer that seems plausible, at least in some examples. Therefore, either this comment on the help page is wrong (or obsolete) or the answer can not be trusted with NAs.

Author(s)

Spencer Graves for the ARIMA{FinTS} wrapper for arima, written by the R Core Team, and Box.test, written by A. Trapletti. John Frain provided the citation to a proof in Brockwell and Davis (1990) that the degrees of freedom for the approximating chi-square distribution of the Ljung-Box statistic should be adjusted for the number of parameters estimated. Michal Miklovic provided the citation to Enders (2004).

References

Brockwell and Davis (1990) Time Series: Theory and Methods, 2nd Edition (Springer, page 310).

Walter Enders (2004) Applied Econometric Time Series (Wiley, pp. 68-69)

Greta Ljung and George E. P. Box (1978) 'On a measure of lack of fit in time series models', Biometrika, vol. 66, pp. 67-72.

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 2)

Patrick Burns (2002) 'Robustness of the Ljung-Box Test and its Rank Equivalent', https://www.burns-stat.com/pages/Working/ljungbox.pdf, accessed 2007.12.29.

See Also

arima Box.test tsdiag

Examples

##
## Examples from 'arima'
##
lh100 <- ARIMA(lh, order = c(1,0,0))
lh100$Box.test
# df = 3 = round(log(lh)) - 1
# 2 parameters are estimated, but 1 is 'intercept',
# so it doesn't count in the 'df' computation

lh500 <- ARIMA(lh, order = c(5,0,0))
lh500$Box.test
# round(log(length(lh))) = 4
# Default Box.test.lag = min(5+1, 4) = 6,
# so df = 1;  without the min(5+1, ...), it would be -1.
lh500$Box.test$method # lag = 6

lh101 <- ARIMA(lh, order = c(1,0,1))
lh101$Box.test
# works with mixed ARMA

USAccD011 <- ARIMA(USAccDeaths, order = c(0,1,1),
                   seasonal = list(order=c(0,1,1)))
USAccD011$Box.test
# df = round(log(length(USAccDeaths))) - 2:
# correct 'df' with nonstationary 'seasonal' as well

LakeH200 <- ARIMA(LakeHuron, order = c(2,0,0),
                  xreg = time(LakeHuron)-1920)
LakeH200$Box.test
# correct 'df' with 'xreg'
LakeH200$r.squared

## presidents contains NAs
## graphs in example(acf) suggest order 1 or 3
require(graphics)
(fit1 <- ARIMA(presidents, c(1, 0, 0)))
fit1$Box.test
tsdiag(fit1)
##
## Example with multiple 'xreg' variables
##
tLH <- as.numeric(time(LakeHuron)-1920)
tLH2 <- cbind(timeLH.1920 = tLH, time.sq = tLH*tLH)

LakeH200. <- ARIMA(LakeHuron, order=c(2,0,0), xreg=tLH2)
LakeH200.$r.squared

Autocorrelation Function

Description

Plot the ACF without the traditional noninformation unit spike at lag 0.

Usage

Acf(x, lag.max = NULL, type = c("correlation", "covariance", "partial"),
    plot = TRUE, na.action = na.fail, demean = TRUE, ...)

## S3 method for class 'Acf'
plot(x, ci = 0.95, type = "h", xlab = "Lag", ylab = NULL, ylim = NULL, 
     main = NULL, ci.col = "blue", ci.type = c("white", "ma"), 
     max.mfrow = 6, ask = Npgs > 1 && dev.interactive(), 
     mar = if (nser > 2) c(3, 2, 2, 0.8) else par("mar"), 
     oma = if (nser > 2) c(1, 1.2, 1, 1) else par("oma"), 
     mgp = if (nser > 2) c(1.5, 0.6, 0) else par("mgp"), 
     xpd = par("xpd"), cex.main = if (nser > 2) 1 else par("cex.main"), 
     verbose = getOption("verbose"), acfLag0 = FALSE, ...)

Arguments

Details

These functions are provided to make it easy to plot an autocorrelation function without the noninformative unit spike at lag 0. This is done by calling plot(x, acfLag0 = FALSE, ...). Apart from the 'acfLag0' argument, the rest of the arguments are identical to those for 'acf' and 'plot.acf'.

Value

for acf, an object of class 'Acf', which inherits from class 'acf', as described with help('acf', package='stats').

for plot.Acf, NULL

Author(s)

Spencer Graves for the FinTS modification of 'plot.acf'.

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley)

See Also

acf plot.acf Box.test AutocorTest

Examples

data(m.ibm2697)
Acf(m.ibm2697)
Acf(m.ibm2697, lag.max=100)
Acf(m.ibm2697, lag.max=100, main='Monthly IBM returns, 1926-1997')

ARCH LM Test

Description

Lagrange Multiplier (LM) test for autoregressive conditional heteroscedasticity (ARCH)

Usage

ArchTest (x, lags=12, demean = FALSE) 

Arguments

Details

Computes the Lagrange multiplier test for conditional heteroscedasticity of Engle (1982), as described by Tsay (2005, pp. 101-102).

This is provided for compatibility with 'archTest' in the S-Plus script in Tsay (p. 102).

Value

an object of class 'htest'

Author(s)

Bernhard Pfaff

See Also

AutocorTest

Examples

data(m.intc7303)
intcLM <- ArchTest(log(1+as.numeric(m.intc7303)), lag=12)
# Matches answer on Tsay (p. 102) 

Box-Ljung autocorrelation test

Description

Ljung-Box test for autocorrelation

Usage

AutocorTest(x, lag = ceiling(log(length(x))),
            type = c("Ljung-Box", "Box-Pierce", "rank"),
            df = lag )

Arguments

Details

This is provided for compatibility with 'autocorTest' in the S-Plus script in Tsay (p. 30). It is a wrapper for the R function Box.test.

Value

a list of class 'htest' containing the following components:

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley)

Patrick Burns (2002) 'Robustness of the Ljung-Box Test and its Rank Equivalent', https://www.burns-stat.com/pages/Working/ljungbox.pdf, accessed 2007.12.29.

See Also

Box.test ARIMA

Examples

data(m.ibm2697)
AutocorTest(m.ibm2697, 5)

AT4 <- AutocorTest(m.ibm2697, 5, df=4)
str(AT4) # $method = "Box-Ljung test (lag = 5)"

Financial Time Series summary statistics

Description

Summary statistics as in Table 1.2, Tsay (2005), including the start date, number of observations, mean, standard deviation, skewness, excess kurtosis, min and max.

Usage

FinTS.stats(x) 

Arguments

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, p. 11)

See Also

index, sum, is.na, mean, sd, skewness, kurtosis, min, max

Examples

FinTS.stats(rep(1, 5))

data(d.c8603)
FinTS.stats(100*d.c8603[, "C"])

## The following generates an error, because FinTS.stats
## expects a vector of class 'zoo', and d.c8603 is a matrix
#FinTS.stats(100*d.c8603) 

List of the names of files downloaded from the "Analysis of Financial Data" web site.

Description

A list organized by chapter and text vs. exercises of the files downloaded from the web site associated with Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Springer) and stored in "~library/FinTS/scripts/TsayFiles". These facilitate the process of creating and updating the 'FinTS' package (and documenting the creation process).

Usage

TsayFiles

Format

TsayFiles

A list with names 'ch01', 'ch02', ..., 'ch12' for components describing the files associated with the corresponding chapter.

Each chapter component is a list with 'text' and 'exercises' components, where 'text' and 'exercises' are each a character array giving names for 'data', 'file', 'url', and 'found' for the data referenced in the text or exercises of that chapter:

data

'file' without the extension, e.g, 'd-ibmvwewsp6203' for daily simple returns of IBM, VW, EW, SP (7/3/62-12/31/03) = 'file' without the extension.

file

short file name = 'data' plus the extension, e.g., 'd-ibmvwewsp6203'.txt' for daily simple returns of IBM, VW, EW, SP (7/3/62-12/31/03)

url

universal resource locator for the data, e.g.,

"http://faculty.chicagogsb.edu/ruey.tsay/teaching/fts2/d-ibmvwewsp6203.txt"

found

'TRUE' if the data were found, 'FALSE' if the attempt to access the url failed.

NOTES:

(1) 13 files are referenced twice, and 2 are referenced three times on the web page. This redundancy is retained in 'TsayFiles'.

(2) A few files (most noticably some with with '.dat' extension) are referenced in the HTML code without an apparent visible link. These 'invisible files' are retained in 'TsayFiles'.

FinTS.url

A character string giving the universal resource locator (URL) associated with the Tsay (2005) book:

FinTS.url <- "https://faculty.chicagobooth.edu/ruey-s-tsay/teaching"

Source

Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Springer)

Examples

data(TsayFiles)
TsayFiles$ch01$exercises

Asymptotic PCA

Description

Asymptotic Principal Components Analysis for a fixed number of factors

Usage

apca(x,nf)

Arguments

Details

NOTE: This is a preliminary version of this function, and it may be modified in the future.

Value

A list with four components:

Author(s)

Ruey Tsay

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, sec. 9.6, pp. 436-440)

See Also

princomp

Examples

# Consider the monthly simple returns of 40 stocks on NYSE and NASDAQ
# from 2001 to 2003 with 36 observations.
data(m.apca0103)
dim(m.apca0103)
M.apca0103 <- with(m.apca0103,
    array(return, dim = c(36, 40),
          dimnames = list(as.character(date[1:36]),
                          paste("Co", CompanyID[seq(1, 1440, 36)], sep = ""))))

# The traditional PCA is not applicable to estimate the factor model
# because of the singularity of the covariance matrix.  The asymptotic
# PCA provides an approach to estimate factor model based on asymptotic
# properties. For the simple example considered, the sample size is
# $T$ = 36 and the dimension is $k$ = 40. If the number of factor is
# assumed to be 1, the APCA gives a summary of the factor loadings as
# below:
#
apca40 <- apca(M.apca0103, 1)
#
# (min, 1st Quartile, median, mean, 3rd quartile, max) =
# (0.069, 0.432, 0.629, 0.688, 1.071, 1.612).
#
# Note that the sign of any loading vector is not uniquely determined
# in the same way as the sign of an eigenvector is not uniquely
# determined. The output also contains the summary statistics of the
# R-squares of individual returns, i.e. the R-squares measuring the
# total variation of individual return explained by the factors. For
# the simple case considered, the summary of R-squares is (min, 1st
# Quartile, median, mean, 3rd quartile, max) =
# (0.090,0.287,0.487,0.456,0.574,0.831).

Conditionally convert x to yearmon if the conversion is unique, retaining x as names.

Description

Convert x to class "yearmon". If duplicate months are found, return x. Otherwise, return the conversion with names = x.

Usage

as.yearmon2(x, ...)

Arguments

Details

Dates for some monthly data include the day of the month on which the data were published. For many purposes, one would like to have the data as a zoo object with a yearmon index, while still retaining the full date for other purposes.

If the yearmon form of the input is not unique, as.yearmon2 returns the input unchanged with a warning. Otherwise, it returns the yearmon conversion with the input as names.

Value

Returns either its argument or its argument converted to class yearmon with names.

See Also

yearmon

Examples

x1 <- as.Date(c("2000-01-01", "2000-01-01"))
as.yearmon2(x1)
#Warning message:
#In as.yearmon2(x1) :
#  1 duplicate months found in 'x';  returning 'x' unchanged

x2 <- as.Date(c("2000-01-01", "2000-02-01"))
as.yearmon2(x2)
# month of x2 with names x2

Financial time series for Tsay (2005, ch. 1)

Description

Financial time series used in examples in chapter 1.

Usage

data(d.ibmvwewsp6203)
data(d.intc7303)
data(d.3m6203)
data(d.msft8603)
data(d.c8603)
data(m.ibmvwewsp2603)
data(m.intc7303)
data(m.3m4603)
data(m.msft8603)
data(m.c8603)
data(m.gs10)
data(m.gs1)
data(d.fxjp00)
data(m.fama.bond5203)
data(m.gs3)
data(m.gs5)
data(w.tb3ms)
data(w.tb6ms)

Format

Objects of class zoo giving simple returns for each trading period (day, week or month) for different periods, with different start dates but typically running to the end of 2003.

d.ibmvwewsp6203, m.ibmvwewsp2603

Zoo objects with 4 columns (IBM, VW, EW, and SP). Daily data starts with 1962-07-03. Monthly data starts with 1926-01-30.

d.intc7303, m.intc7303

Matrices of class zoo with a single column "Intel" starting from January 1973.

d.3m6203, m.3m6203

Matrices of class zoo with a single column "MMM". Daily data starts with 1962-07-03. Monthly data starts with 1946-02-28.

d.msft8603, m.msft8603

Matrices of class zoo with a single column "MSFT" starting from 1906-03-14.

d.c8603, m.c8603

Matrix of class zoo with a single column "C" starting from 1986-10-30.

m.gs10, m.gs1

Monthly 10-yr and 1-yr Treasury constant maturity rates (4/53-3/04)

d.fxjp00

Daily exchange rate between U.S. dollar and Japanese yen

m.fama.bond5203

Monthly bond returns as follows:

m1.12

1-12m

m24.36

24-36m

m48.60

48-60m

m61.120

61-120m

m.gs3, m.gs5

Monthly 3-yr and 5-yr Treasury constant maturity rates

w.tb3ms, w.tb6ms

Weekly Treasury Bill rates

Details

The first 16 of these objects contain daily and monthly simple returns for 8 financial time series analyzed Tsay (2005, Table1.2). These 8 are SP (Standard & Poors), EW, IBM, Intel, Microsoft, and Citi-Group, beginning at different times and running to the end of 2003.

The others are used elsewhere in chapter 1.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 1)

See Also

FinTS.stats

Examples

# First half of Table 1.2:
data(d.ibmvwewsp6203)
data(d.intc7303)
data(d.3m6203)
data(d.msft8603)
data(d.c8603)
(Daily.Simple.Returns.pct <- rbind(
    SP = FinTS.stats(100*d.ibmvwewsp6203[, "SP"]),
    VW = FinTS.stats(100*d.ibmvwewsp6203[, "VW"]),
    EW = FinTS.stats(100*d.ibmvwewsp6203[, "EW"]),
    IBM= FinTS.stats(100*d.ibmvwewsp6203[, "IBM"]),
    Intel=FinTS.stats(100*d.intc7303[, "Intel"]),
    MMM= FinTS.stats(100*d.3m6203[, "MMM"]),
    MSFT=FinTS.stats(100*d.msft8603[, 'MSFT']),
    C  = FinTS.stats(100*d.c8603[, "C"])
) )

(Daily.log.Returns.pct <- rbind(
    SP = FinTS.stats(100*log(1+d.ibmvwewsp6203[, "SP"])),
    VW = FinTS.stats(100*log(1+d.ibmvwewsp6203[, "VW"])),
    EW = FinTS.stats(100*log(1+d.ibmvwewsp6203[, "EW"])),
    IBM= FinTS.stats(100*log(1+d.ibmvwewsp6203[, "IBM"])),
    Intel=FinTS.stats(100*log(1+d.intc7303[,"Intel"])),
    MMM= FinTS.stats(100*log(1+d.3m6203[, "MMM"])),
    MSFT=FinTS.stats(100*log(1+d.msft8603[, 'MSFT'])),
    C  = FinTS.stats(100*log(1+d.c8603[, "C"]))
) )

Financial time series for Tsay (2005, ch. 2)

Description

Financial time series used in examples in chapter 2.

Usage

data(m.ibm2697)
data(m.vw2697)
data(q.gnp4791)
data(m.ibm3dx2603)
data(m.3m4697)
data(q.gdp4703)
data(d.sp9003lev)
data(q.jnj)
data(m.decile1510)
data(w.gs1n36299)

Format

Objects of class zoo giving simple returns for each trading period (day, week or month) for different periods.

m.ibm2697, m.vw2697

Monthly returns for IBM stock and the value weighted index from 1926 to 1997.

q.gnp4791

Growth rate of U.S. quarterly real gnp, from 1947Q2 to 1991Q1.

m.ibm3dx2603

Monthy returns of IBM stock, the value and equal weighted and Standard and Poors indices from 1926 through 2003.

m.3m4697

Monthly simple returns of 3M stock from Feb., 1946 through Dec. 2003.

q.gdp4703

U.S. quarterly GDP from 1947 through 2003

d.sp9003lev

Daily values of S&P 500 index from 1990 through 2003.

q.jnj

Quarterly earnings of Johnson & Johnson from 1960 through 1980.

m.decile1510

Monthly simple returns of Deciles 1, 5, 10. Decile 1 means the weighted returns of companies in the first 10 percent of market cap (i.e. 0 to 10). (Thus, it is not the 10th percentile.) Decile 10 means the returns of the top 10 percent of the companies (market cap). Therefore, decile 1 is the smallest listed companies, and decile 10 is for the largest companies.

The 'index' of 'm.decile1510' has class 'Date'. Since it's a monthly series, it would be better for many purposes if it had 'index' of class 'yearmon'. See the 'examples' below for how to achieve this conversion.

w.gs1n36299

zoo object with two columns, 'gs1' and 'gs3', giving weekly 1-yr & 3-yr interest rates from 1962-01-05 through 2007-11-02. These data were reextracted from the Federal Reserve Bank at St. Louis to replace data from the book's web site that had obvious data quality problems (e.g., a date of 1962-08-32).

To get data covering January 4, 1962, through September 10, 1999, use

window(w.gs1n36299, start = as.Date("1962-01-12"), end=as.Date("1999-09-10"));

see 'examples' below.

Author(s)

Spencer Graves with help from Gabor Grothendieck.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 2)

See Also

ch01data

Examples

## m.decile1510 has 'index' of class 'Date'
## Since it's a monthly series, for many purposes,
## it should have 'index' of class 'yearmon'.
## To get this, do the following:
##
data(m.decile1510)
mDecile1510 <- zoo(m.decile1510, as.yearmon(index(m.decile1510)))

## w.gs1n36299 covers a broader range than used in
## Tsay (2005, sec. 2.9, pp. 80ff):  subset using 'window':
##
data(w.gs1n36299)
w.gs1n3 <- window(w.gs1n36299, start = as.Date("1962-01-12"),
    end = as.Date("1999-09-10"))

Financial time series for Tsay (2005, ch. 3)

Description

Financial time series used in examples in chapter 3.

Usage

#m.intc7303
data(exch.perc)
data(sp500)
#m.ibm2697
#d.ibmvwewsp6203
data(m.ibmspln)
data(m.ibmsplnsu)
data(d.sp8099)

Format

Three data stes used in chapter 3 are also used in chapter 1 or 2 and are documented with 'ch01data' or 'ch02data': In particular, 'm.intc7303' and 'd.ibmvwewsp6203' are used in chapters 1 and 3 and are documented with 'ch01data'; 'm.ibm2697' is used in chapters 2 and 3 is documented with ch02data.

The other data sets used in chapter 3 are as follows:

exch.perc

numeric vector of length 2497 giving percentage changes in the exchange rate between the German mark and the US dollar in 10 minute intervals, June 5-19, 1989. (The book describes analyses of 2488 observations. If these 2497 observations are plotted, it is difficult to see any differences from Figure 3.2.)

sp500

object of class 'zooreg' giving the monthly excess returns of the S&P 500 index starting from 1926. This zooreg object is labeled assuming it starts in January, though the book does not say whether it starts in January or just some time in 1926. (Many of the files included date with the data, but 'sp500.dat' did not.)

m.ibmspln

object of class 'zooreg' giving the monthly log returns of IBM stock and S&P 500 index from January 1926 to December 1999 for 888 observations. NOTE: The examples in the book use only the first 864 of these observations.

m.ibmsplnsu

same as 'm.ibmspln' but with a third column 'summer' that is 1 in June, July and August, and 0 otherwise.

d.sp8099

zoo object giving the average daily returns of the S&P 500 from 1980 through 1999.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 3)

See Also

ch01data, ch02data

Financial time series for Tsay (2005, ch. 4)

Description

Financial time series used in examples in chapter 4.

Usage

data(m.unrate)
#d.ibmvwewsp6203
#m.3m4697
#q.gnp4791
data(w.3mtbs7097)
data(m.ibmln2699)
data(q.unemrate)

Format

Three data stes used in chapter 4 are also used earlier: 'd.ibmvwewsp6203' is used in chapter 1, and 'm.3m4697' and 'q.gnp4791' are used in chapter 2; these three data objects are docuemented in 'ch01data' or 'ch02data'.

The other data sets used in chapter 4 are as follows:

m.unrate

zoo object giving the monthly US civilian unemployment rate from 1948 through 2004.

w.3mtbs7097

zoo object giving the US weekly 3-month treasury bill rate in the secondary market from 1970 through 1997.

m.ibmln2699

zoo object giving the monthly log returns in percentages of IBM stock from 1926 through 1999.

q.unemrate

zoo object giving the US quarterly unemployment rate seasonally adjusted from 1948 through 1993.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 4)

See Also

ch01data ch02data

Financial time series for Tsay (2005, ch. 5)

Description

Financial time series used in examples in chapter 5.

Usage

data(ibm)
data(ibm9912.tp)
data(ibmdurad)
data(ibm1to5.dur)
data(ibm91.ads)
data(ibm91.adsx)
data(day15.ori)
data(day15)

Format

ibm

IBM transactions data (11/1/1990 - 1/31/1991)

data.frame of date.time, volume, bid, ask, and price of IBM stock transactions. date.time is of class 'chron', while volume, bid, ask, and price are all numeric. Some tranactions have the same date.time values, which is why this is a data.frame and not a zoo object.

ibm9912.tp

IBM transactions data of December 1999: data.frame of date.time and price.

ibmdurad

Adjusted time durations between trades of IBM stock (11/01/1990-1/31/1991).

Format: data.frame with columns date.time and adjusted.duration

ibm1to5.dur

subset of 'ibmdurad' limited to positive durations in the first 5 trading days.

ibm91.ads

a data.frame on the changes in the price of IBM stock transactions between November 1, 1990 and January 31, 1991. This period includes 63 trading days, during which 59,838 transactions were recorded during normal trading hours. The first transaction for each day was dropped leaving the 59,775 transactions in this data.frame.

A.priceChange

1 if a price change from the previous trade, 0 otherwise

DirectionOfChg

1 if positive, -1 if negative, 0 if no change

SizeInTicks

Size of the price change in number of ticks of 1/8 of a US dollar.

NOTE: In this data.frame there are 10 anomalous records for which A.priceChange !=0 but SizeInTicks == 0. These correspond to price changes of half a tick, which got rounded down to 0.

ibm91.adsx

a data.frame with 6 variables the same transactions as in 'ibm91.ads':

volume.thousands

thousands of shares traded

time.betw.trades

seconds between the previous two trades

bid.ask.spread

the bid-ask spread in USD of the current transaction.

A.priceChange

1 if the previous trade involved a price change from its predacessor, and 0 otherwise

DirectionOfChg

1 if the previous change was positive, -1 if negative, 0 if no change

SizeInTicks

Size of the price change in the previous trade in number of ticks of 1/8 of a US dollar. NOTE: The last three columns are ibm91.ads lagged one transaction, so ibm91.adsx[-1, 4:5] == ibm91.ads[-59775, ], with 24 exceptions.

day15.ori

data.frame with the transaction time and the stock price for the 728 IBM stock transactions that occurred during normal trading hours on November 21, 1990.

day15

a zoo object with the following columns supposedly summarizing only the price changes in day15.ori:

timeBetwPriceChg

time in seconds since the last price change

DirectionOfChange

1 if the price increased, -1 if it decreased

priceChgTicks

price change in number of ticks of USD 1/8.

nTradesWoChg

number of trades without a price change since the previous price change ... supposedly. These numbers do not match a manual extraction of these data from 'day15.ori'.

multTrans

1 if there were multiple transactions within the same one second interval, 0 if not.

dailyCumChg

cumulative price change in USD since the start of normal trading on November 21, 1990.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 5)

See Also

ch01data ch02data

Financial time series for Tsay (2005, ch. 6)

Description

Financial time series used in examples in chapter 6.

Usage

data(d.ibmy98)
data(d.cscoy99)

Format

Objects of class zoo giving returns for each trading day for different periods

d.ibmy98

Zoo object giving daily simple returns of IBM stock for each trading day in 1998.

d.cscoy99

Zoo object giving daily log returns of Cisco stock for each trading day in 1999.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 6)

See Also

ch01data, ch02data, ch03data, ch04data, ch05data

Financial time series for Tsay (2005, ch. 7)

Description

Financial time series used in examples in chapter 7.

Usage

data(d.ibm6298wmx)
data(d.intc7297)

Format

d.ibm6298wmx

a zoo object of 9190 observations on several series relating to IBM stock, 1962-07-03 to 1998-12-31:

dailySimpleRtns

daily simple returns in percentages of IBM stock

day

numbers 1:9190

meanCorrectedLogRtns

mean-corrected log returns

Q4

1 for October, November, December, and 0 otherwise

drop2.5pct

an indicator variable for the behavior of the previous trading day. Specifically, this is 1 if the meanCorrectedLogRtns for the previous day was at most (-0.025).

nOfLast5outside2.5pct

number of the last 5 days for which the meanCorrectedLogRtns exceeded +/-2.5

annualTrend

an annual trend defined as (year-1961)/38.

GARCH1.1volatility

a volatility series based on a Gaussian GARCH(1,1) model for the mean-corrected log returns.

The simpleDailyRtns and the zoo index are from 'd-ibm6298.txt' from the book's web site.

The 'day' and 'meanCorrectedLogRtns' are from 'd-ibmln98wm.txt'.

The last 5 columns are from 'd-ibml25x.txt'; they are described on p. 332 of the book.

d.intc7297

a zoo object of daily log returns of Intel stock, 1972-12-15 to 1997-12-31.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 7)

See Also

ch01data, ch02data, ch03data, ch04data, ch05data, ch06data

Financial time series for Tsay (2005, ch. 8)

Description

Financial time series used in examples in chapter 8.

Usage

data(m.ibmsp2699ln)
data(m.bnd)
data(m.gs1n3.5301)
data(w.tb3n6ms)
data(sp5may)

Format

m.ibmsp2699ln

zoo object giving monthly simple and log returns of IBM stock and the Standard and Poor's 500 from 1926 through 1999. (This combines files 'm-ibmsp2699.txt' and 'm-ibmspln.txt' from the book's web site.)

m.bnd

zoo object giving the monthly simple returns of 30, 20, 10, 5 and 1 year maturity bonds from 1942 through 1999.

m.gs1n3.5301

zoo object giving 1 and 3 year US Treasury constant maturity interest rates from April 1953 to January 2001 (used in Example 8.6, pp. 373ff).

w.tb3n6ms

zoo object giving weekly 3 and 6 month US Treasury Bill interest rates from 1958-12-12 to 2004-08-06 (used in Sect. 8.6.5, pp. 385ff).

sp5may

A data.frame of 7061 observations on 4 variables based on minute-by-minute observations of the Standard and Poor's 500 Futures and prices in May 1993.

These data are used, after some processing, in Tsay(Sect. 8.7.2, pp. 392ff). Unfortunately, it's not yet clear what these numbers are. The following is a current guess and will doubtless change in the future.

logFuture

logarithms of June Futures contracts traded at the Chicago Mercantile Exchange. The first difference of this series appears to be plotted in Figure 8.16(a), after replacing '10 extreme values (5 on each side) by the simple average of their two nearest neighbors.' (p. 392)

logPrice

logarithms of Standard and Poor's 500 price levels. The first differences of this series appears to be plotted in Figure 8.16(b), after adjustment similar to that for 'logFuture'.

dailyAvgSomething

numbers that assume 19 distinct levels separated by 18 discrete jumps. The name of this will likely change whenever more information about it can be obtained for this documentation.

day

index for the 19 distinct levels assumed by 'dailyAvgSomething'. This is probably the trading day in May 1993. However, there appear to have been 20 trading days in that month, so if these 19 levels do correspond to trading days, it's not clear which date is missing.

These data were analyzed by Forbes, Kalb, and Kofman (1999); Tsay (1998) was also referenced with the discussion of the analysis of these data.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Forbes, C. S., Kalb, G. R. J., and Kofman, P. (1999) 'Bayesian Arbitrage Threshold Analysis', Journal of Business and Economic Statistics, 17: 364-372.

Ruey Tsay (1998) 'Testing and Modeling Multivariate Threshold Models', Journal of the American Statistical Association, 93: 1188-1202.

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 8)

See Also

ch01data, ch02data

Financial time series for Tsay (2005, ch. 9)

Description

Financial time series used in examples in chapter 9.

Usage

data(m.fac9003)
data(m.cpice16.dp7503)
data(m.barra.9003)
data(m.5cln)
#data(m.bnd) <- documented with ch08, also used in ch09
data(m.apca0103)

Format

m.fac9003

a zoo object of 168 observations giving simple excess returns of 13 stocks and the Standard and Poor's 500 index over the monthly series of three-month Treasury bill rates of the secondary market as the risk-free rate from January 1990 to December 2003. (These numbers are used in Table 9.1.)

AA

Alcoa

AGE

A. G. Edwards

CAT

Caterpillar

F

Ford Motor

FDX

FedEx

GM

General Motors

HPQ

Hewlett-Packard

KMB

Kimberly-Clark

MEL

Mellon Financial

NYT

New York Times

PG

Proctor & Gamble

TRB

Chicago Tribune

TXN

Texas Instruments

SP5

Standard & Poor's 500 index

m.cpice16.dp7503

a zoo object of 168 monthly on two macroeconomic variables from January 1975 through December 2002 (p. 412):

CPI

consumer price index for all urban consumers: all items and with index 1982-1984 = 100

CE16

Civilian employment numbers 16 years and over: measured in thousands

m.barra.9003

a zoo object giving monthly excess returns of ten stocks from January 1990 through December 2003:

AGE

A. G. Edwards

C

Citigroup

MWD

Morgan Stanley

MER

Merrill Lynch

DELL

Dell, Inc.

IBM

International Business Machines

AA

Alcoa

CAT

Caterpillar

PG

Proctor & Gamble

m.5cln

a zoo object giving monthly log returns in percentages of 5 stocks from January 1990 through December 1999:

IBM

International Business Machines

HPQ

Hewlett-Packard

INTC

Intel

MER

Merrill Lynch

MWD

Morgan Stanley Dean Witter

m.apca0103

data.frame of monthly simple returns of 40 stocks from January 2001 through December 2003, discussed in sect. 9.6.2, pp. 437ff.

CompanyID

5-digit company identification code

date

the last workday of the month

return

in percent

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 7)

See Also

ch01data, ch02data, ch03data, ch04data, ch05data, ch06data

Examples

data(m.apca0103)
dim(m.apca0103)
# 1440 3;  1440 = 40*36
# Are the dates all the same?
sameDates <- rep(NA, 39)
for(i in 1:39)
    sameDates[i] <- with(m.apca0103,
                         all.equal(date[1:36], date[(i*36)+1:36]))
stopifnot(all(sameDates))
M.apca0103 <- with(m.apca0103, array(return, dim = c(36, 40), dimnames =
    list(NULL, paste("Co", CompanyID[seq(1, 1440, 36)], sep=""))))

Financial time series for Tsay (2005, ch. 10)

Description

Financial time series used in examples in chapter 10.

Usage

data(d.hkja)
data(m.pfe6503)
data(m.mrk6503)
#data(m.ibmsp2699)
# <- 2 of the 4 columns in m.ibmsp2699ln
# documented with ch08data
data(d.spcscointc)

Format

One data set used in chapter 10 is also used earlier: 'm.ibmsp2699' is the first 2 of the 4 columns of 'm.ibmsp2699ln' used in chapter 8.

The other data sets used in chapter 10 are as follows:

d.hkja

zoo object giving the daily log returns of HK and Japan market indices from 1996-01-01 through 1997-05-05 (used in Example 10.1).

m.pfe6503, m.mrk6503

zoo objects giving the monthly simple returns including dividends of Pfizer and Merk stocks.

d.spcscointc

data.frame giving 2275 daily log returns of three items from January 2, 1991 through December 31, 1999:

SP500

Standard & Poor's 500 index

Cisco

Cisco stock

Intel

Intel stock

NOTE: This date range seems to include 2280 trading days in the New York Stock Exchange. Since the file on the book's web site did not include dates and since there appear to be more trading days than observations, dates are not currently provided with these observations. This may change with a future revision of this package.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 10)

See Also

ch01data, ch02data, ch03data, ch04data, ch05data, ch06data, ch07data, ch08data, ch09data

Financial time series for Tsay (2005, ch. 11)

Description

Financial time series used in examples in chapter 11.

Usage

data(aa.3rv)
# m.fac9003 described in ch09data
# q.jnj described in dh02data

Format

The text of chapter 11 considers one data set not used in previous chapters plus two that are. Monthly excess returns of GM stock are used in Table 9.1 of Chapter 9. Quarterly earnings of Johnson and Johnson are used in Chapter 2.

The data set introduced with Chapter 11 is as follows:

aa.3rv.

a zoo object of daily 5, 10, and 20 minute realized volatility of Alcoa stock from 2003-01-02 through 2004-05-07.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 11)

See Also

ch09data, ch02data

Financial time series for Tsay (2005, ch. 12)

Description

Financial time series used in examples in chapter 12.

Usage

data(w.gs3n1c)
data(w.gs3c)
data(m.sp6299)
data(m.ibmspln6299)
data(m.sp5.6204)
data(m.geln)

Format

w.gs3n1c

a zoo object of the change series of weekly US interest rates (3 and 1 year maturities) from Jan. 5, 1962, to Sep. 10, 1999. This was obtained via

diff(window(w.gs1n36299, start=as.Date("1962-01-05"), end=as.Date("1999-09-10")))[, 2:1]

to get the dates with the data. Then 'all.equal' confirmed that these numbers matched those in the file read from the web site (which did not have dates).

These are used in Example 12.1, pp. 556ff.

w.gs3c

a zoo object giving the change series of weekly US 3-year maturity interest rates from March 18, 1988, to Sept. 10, 1999. This was obtained via

window(w.gs3n1c[, 1], start=as.Date("1988-03-18"), end = as.Date("1999-09-10")).

Then 'all.equal' confirmed that these numbers matched those read from the web site.

These data are used in Example 12.2, pp. 564ff.

m.sp6299

Monthly log returns of S&P 500 index from January 1962 to December 1999. These data are used in Example 12.3, pp. 569ff.

These data are a subset of 'm.ibmspln', used in chapter 3. That series has dates, which were not provided in the file associated with this series on the book's web site. Moreover, the file with chapter 12 has only 4 significant digits where the earlier file has 6. Since the other data are otherwise identical, this 'm.sp6299' was constructed as

window(m.ibmspln[, 2], start = yearmon(1962), end = yearmon(1999+11/12)).

m.ibmspln6299

Monthly log returns of IBM stock and the S&P 500 index from January 1962 to December 1999. These data are used in Example 12.4, pp. 573ff.

These data are an expansion of 'm.sp6299' and were similarly obtained from 'm.ibmspln'.

m.sp5.6204

Monthly log returns of S&P 500 index from January 1962 to November 1999. These data are used in Example 12.5, pp. 586ff.

m.geln

Monthly log returns of GE stock from January 1926 to December 1999. These data are used in Example 12.6, pp. 591ff.

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 12)

See Also

ch01data, ch02data, ch03data, ch04data, ch05data, ch06data, ch07data, ch08data, ch09data, ch10data, ch11data

Compute compound interest

Description

Compute compound interest for a given number of periods, compounding with an indicated frequency per period.

Usage

compoundInterest(interest, periods = 1, frequency = 1, net.value = FALSE)

simple2logReturns(R)

Arguments

Details

These functions are vectorized for all arguments. The code uses optionally expm1(x) = exp(x) - 1 and log1p(x) = log(1+x) which can preserve numerical precision for x very close to 0.

Value

vector of the length of the longest argument.

References

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, p. 6)

Examples

# "Net Value" column of Tsay Table 1.1, p. 4
compoundInterest(0.1, frequency = c(1, 2, 4, 12, 52, 365, Inf),
                      net.value = FALSE)
# Example 1.1, p. 6
compoundInterest(.0446, freq = Inf)
# Inverse of Example 1.1
simple2logReturns(.0456)

Find complex conjugate pairs

Description

Find all complex conjugate pairs in a vector of complex numbers and return one number from each pair.

Usage

findConjugates(x, complex.eps = .Machine[["double.eps"]])

Arguments

Details

1. Compute normalization m2 = outer(abs(x), abs(x), max).

2. Compute complex differences c2 = abs(outer(x, Conj(x), "-")) / m2.

3. If any abs(c2) < complex.eps, make sure the numbers are not duplicate reals via

(d2 = abs(outer(x, x, "-"))) > complex.eps.

Value

a complex vector with one representative of each complex pair found

Author(s)

Spencer Graves and Ravi Varadhan

See Also

plotArmaTrueacf

Examples

# none
findConjugates(NULL)
findConjugates(numeric(0))
findConjugates(0:4)
findConjugates(rep(0:1,each=3))

# some
findConjugates(c(1+1i, 0, 1-1i, 2-2i, 3, 2+2i, NA))

# Testing with polyroot and solve(polynomial(...))
set.seed(1234)
if(require(polynom)){
  p <- polynomial(sample(1:10, size=45, rep=TRUE)) # degree 44
  z <- solve(p)
  findConjugates(z, complex.eps=.Machine$double.eps)
  # this identifies all 21 conjugate pairs, R 2.6.0 for Windows

  z1 <- polyroot(p)
  findConjugates(z1, complex.eps=.Machine$double.eps)
  # this only identifies only 3 conjugate pairs, R 2.6.0 for Windows
}

Directory of a package

Description

Display a partial or complete directory of a package. By default, suppress common package contents to focus on 'demo', 'doc', 'scripts', and similar subdirectories whose contents might contain examples that could make it easier to learn capabilities of the package.

Usage

package.dir(package = 'base', lib.loc = NULL,
      exclude = c('chtml', 'data', 'help', 'html', 'latex', 'libs',
                  'man', 'Meta', 'po', 'R', 'R-ex', 'src'),
      include = NULL, pattern = NULL, recursive = FALSE)

Arguments

Details

1. fullPath <- system.file(package = package, lib.loc = lib.loc)

2. Dir <- dir(fullPath)

3. Restrict Dir only to 'include' if provided and to all but 'exclude' otherwise.

4. If recursive, return a list produced by dir for each of the subdirectories of interest determined in step 3. Else, return only the list of subdirectories from step 3.

Value

If recursive, a list of the contents of the subdirectories of interest. Else, a character vector of the names of the relevant subdirectories.

Author(s)

Spencer Graves

See Also

system.file dir file.path

Examples

package.dir() # 'demo'
package.dir(recursive = TRUE) # contents of 'demo'
package.dir('nlme') # 'mlbook', 'scripts'

Plot loadings

Description

Plots loadings as a separate barplot for each factor.

Usage

## S3 method for class 'loadings'
plot(x, n = 5, k = ncol(x), mfrow = c(k, 1), ...)

Arguments

Details

The top n components of each of the top k factors are displayed in a separate barplot.

Value

Return value is a list of the return values from each barplot invocation.

See Also

barplot

Examples

data(m.barra.9003)
rtn <- m.barra.9003
stat.fac <- factanal(rtn, factors = 3)
m.barra.loadings <- loadings(stat.fac)
plot(m.barra.loadings)

Plot the theoretical ACF corresponding to an ARMA model

Description

Compute the roots and theoretical ACF corresponding to an ARMA model

Usage

plotArmaTrueacf(object, lag.max = 20, pacf = FALSE, plot = TRUE,
          xlab = "lag", ylab = c("ACF", "PACF")[1+pacf],
          ylim = c(-1, 1) * max(ACF), type = "h",
          complex.eps = 1000 * .Machine[["double.neg.eps"]], ...)

Arguments

Details

1. Compute and test stationarity. An ARMA process is stationary if all the roots of its AR component lie inside the unit circle (Box and Jenkins, 1970). If the process is not stationary, a warning is issued, and no plot is produced.

2. Compute and plot the theoretical ACF.

3. Analyze periodicity of any complex roots

Value

a list with the following components

Source

https://faculty.chicagobooth.edu/ruey-s-tsay/teaching

References

George E. P. Box and Gwilym M. Jenkins (1970) Time Series Analysis, Forecasting and Control (Holden-Day, sec. 3.4.1. Stationarity and invertibility properties of Mixed autoregressive-moving average processes)

Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley, ch. 2)

See Also

solve.polynomial, ARMAacf

Examples

# Tsay, Figure 2.3
op <- par(mfcol = c(1, 2))
plotArmaTrueacf(.8, lag.max = 8)
title("(a)")
plotArmaTrueacf(-.8, lag.max = 8)
title("b")
par(op)

# Tsay, Figure 2.4
op <- par(mfrow = c(2,2))
plotArmaTrueacf(c(1.2, -.35))
title("(a)")
plotArmaTrueacf(c(.6, -.4))
title("(b)")
plotArmaTrueacf(c(.2, .35))
title("(c)")
plotArmaTrueacf(c(-.2, .35))
title("(d)")
par(op)

# Tsay, Example 2.1
data(q.gnp4791)
(fit.ar3 <- ar(q.gnp4791, aic = FALSE, order = 3))
plotArmaTrueacf(fit.ar3)

Reading Monthly zoo Series

Description

Read a text file containing monthly data with a date column and return a zoo object with index = a yearmon series with the dates read as names.

Usage

read.yearmon(file, format = "", tz = "", FUN = NULL, regular = FALSE,
             index.column = 1, ...)

Arguments

Details

The arguments are as for read.zoo:

TS <- read.zoo(...)

zoo(coredata(TS), as.yearmon2(index(TS)))

Value

an object of class "zoo" or "zooreg".

See Also

read.table, zoo, read.zoo, coredata, index, as.yearmon2

Examples

## Not run: 
## turn *numeric* first column into yearmon index
## where number is year + fraction of year represented by month
z <- read.zoo("foo.csv", sep = ",", FUN = as.yearmon2)
z2 <- read.yearmon("foo.csv", sep = ",")

## End(Not run)

Run a package script

Description

Run a script associated with a particular chapter

Usage

runscript(x, method = c('run', 'copy', 'view', 'show', 'dir'),
          ask = TRUE, fmt = "ch%02d.R", package = "FinTS",
          subdir = "scripts", lib.loc = NULL)

Arguments

Details

similar to demo or example

Value

the full path and filename, invisibly unless method == 'dir'

Author(s)

Gabor Grothendieck and Spencer Graves

See Also

demo, sprintf, system.file, package.dir, Stangle, vignette, example

Examples

## Not run: 
# provide a menu
runscript()

# run R/library/FinTS/scripts/ch01.R
runscript(1)

# same as:
runscript(1, "run")

# make a copy as 'ch01.R' in the working directory
runscript(1, 'copy')

# display on console only
runscript(1, 'view')

# display using file.show
runscript(1, 'show')

# where is it?
runscript(1, 'dir')

# run R/library/nlme/scripts/afda-ch01.R
if(require(fda))
    runscript(1, fmt = "afda-ch%02d.R", package = "fda")

## End(Not run)

Create local copies of files read from urls

Description

Call 'download.file' with each element of a vector of character strings assumed to be URLs, create a local copy for each, and return a character matrix summarizing what was done.

Usage

url2data(url.) 

Arguments

Details

1. fili <- names(urls.)

2. dati <- fili without its extension, i.e., the part following the last '.'

3. for(i in 1:length(url.)try(download.file(url.[i], fili[i])); found[i] = TRUE if something found and FALSE if not.

4. Return a character matrix with 4 columns: data = dati, file = fili, url = url., and found.

Examples

# See ~R\library\FinTS\scripts\TsayFiles.R