Package 'DAAG'

Description

Various data sets and functions used or referred to in the book Maindonald, J.H. and Braun, W.J. (3rd edn 2010) "Data Analysis and Graphics Using R", plus other selected datasets and functions.

Details

For a list of , use library(help="DAAG").

Author(s)

Author: John H Maindonald

Maintainer: W John Braun [email protected]

Description

Numbers of aberrant crypt foci (ACF) in the section 1 of the colons of 22 rats subjected to a single dose of the carcinogen azoxymethane (AOM), sacrificed at 3 different times.

Usage

ACF1

Format

This data frame contains the following columns:

count

Observed number of ACF in section 1 of each rat colon

endtime

Time of sacrifice, in weeks following injection of AOM

Source

Ranjana P. Bird, Faculty of Human Ecology, University of Manitoba, Winnipeg, Canada.

References

E.A. McLellan, A. Medline and R.P. Bird. Dose response and proliferative characteristics of aberrant crypt foci: putative preneoplastic lesions in rat colon. Carcinogenesis, 12(11): 2093-2098, 1991.

Examples

sapply(split(ACF1$count,ACF1$endtime),var)
plot(count ~ endtime, data=ACF1, pch=16)
pause()
print("Poisson Regression - Example 8.3")
ACF.glm0 <- glm(formula = count ~ endtime, family = poisson, data = ACF1)
summary(ACF.glm0)

# Is there a quadratic effect?
pause()

ACF.glm <- glm(formula = count ~ endtime + I(endtime^2),
  family = poisson, data = ACF1)
summary(ACF.glm)

# But is the data really Poisson?  If not, try quasipoisson:
pause()

ACF.glm <- glm(formula = count ~ endtime + I(endtime^2),
  family = quasipoisson, data = ACF1)
summary(ACF.glm)

Description

These data were collected in a study of how data on various characteristics of the blood varied with sport, body size, and sex of the athlete.

Usage

data(ais)

Format

A data frame with 202 observations on the following 13 variables.

rcc

red blood cell count, in $10^{12} l^{-1}$

wcc

while blood cell count, in $10^{12}$ per liter

hc

hematocrit, percent

hg

hemaglobin concentration, in g per decaliter

ferr

plasma ferritins, ng $dl^{-1}$

bmi

Body mass index, kg $cm^{-2} 10^2$

ssf

sum of skin folds

pcBfat

percent Body fat

lbm

lean body mass, kg

ht

height, cm

wt

weight, kg

sex

a factor with levels f m

sport

a factor with levels B_Ball Field Gym Netball Row Swim T_400m T_Sprnt Tennis W_Polo

Details

Do blood hemoglobin concentrations of athletes in endurance-related events differ from those in power-related events?

Source

These data were the basis for the analyses that are reported in Telford and Cunningham (1991).

References

Telford, R.D. and Cunningham, R.B. 1991. Sex, sport and body-size dependency of hematology in highly trained athletes. Medicine and Science in Sports and Exercise 23: 788-794.

Description

Find the linear transformation which, applied to one set of points in the ($x$, $y$) plane, gives the best match in a least squares sense to a second set of points.

Usage

align2D(lat, long, x1, x2, wts=NULL)

Arguments

Details

Achieves the best match, in a least squares sense, between an ordination and known locations in two-dimensionaL space.

Value

Note

An ordination that is designed to reproduce distances between points is specified only to within an arbitrary rotation about the centroid. What linear transformation of the points ($x1$, $x2$) given by the ordination gives the best match to the known co-ordinates?

Author(s)

John H Maindonald

Examples

if(require(DAAG)&require(oz)){
aupts <- cmdscale(audists)
xy <- align2D(lat = aulatlong$latitude, long = aulatlong$longitude,
              x1 = aupts[, 1], x2 = aupts[, 2], wts = NULL)
oz()
fitcoords <- align2D(lat=aulatlong$latitude,
                      long=aulatlong$longitude,
                      x1=aupts[,1], x2 = aupts[,2],
                      wts=NULL)
x <-with(fitcoords,
         as.vector(rbind(lat, fitlat, rep(NA,length(lat)))))
y <-with(fitcoords,
         as.vector(rbind(long, fitlong, rep(NA,length(long)))))
points(aulatlong, col="red", pch=16, cex=1.5)
lines(x, y, col="gray40", lwd=3)
}

## The function is currently defined as
function(lat, long, x1, x2, wts=NULL){
    ## Get best fit in space of (latitude, longitude)
    if(is.null(wts))wts <- rep(1,length(x1))
    fitlat <- predict(lm(lat ~ x1+x2, weights=wts))
    fitlong <- predict(lm(long ~ x1+x2, weights=wts))
    list(fitlat = fitlat, fitlong=fitlong, lat=lat, long=long)
}

Description

The allbacks data frame gives measurements on the volume and weight of 15 books, some of which are softback (pb) and some of which are hardback (hb). Area of the hardback covers is also included.

Usage

allbacks

Format

This data frame contains the following columns:

volume

book volumes in cubic centimeters

area

hard board cover areas in square centimeters

weight

book weights in grams

cover

a factor with levels hb hardback, pb paperback

Source

The bookshelf of J. H. Maindonald.

Examples

print("Multiple Regression - Example 6.1")
attach(allbacks)
volume.split <- split(volume, cover)
weight.split <- split(weight, cover)
plot(weight.split$hb ~ volume.split$hb, pch=16, xlim=range(volume), ylim=range(weight),
     ylab="Weight (g)", xlab="Volume (cc)")
points(weight.split$pb ~ volume.split$pb, pch=16, col=2)
pause()

allbacks.lm <- lm(weight ~ volume+area)
summary(allbacks.lm)
detach(allbacks)
pause()

anova(allbacks.lm)
pause()

model.matrix(allbacks.lm)
pause()

print("Example 6.1.1")
allbacks.lm0 <- lm(weight ~ -1+volume+area, data=allbacks)
summary(allbacks.lm0)
pause()

print("Example 6.1.2")
oldpar <- par(mfrow=c(2,2))
plot(allbacks.lm0)
par(oldpar)
allbacks.lm13 <- lm(weight ~ -1+volume+area, data=allbacks[-13,])
summary(allbacks.lm13)
pause()

print("Example 6.1.3")
round(coef(allbacks.lm0),2)  # Baseline for changes
round(lm.influence(allbacks.lm0)$coef,2)

Description

Thirty patients were given an anesthetic agent maintained at a predetermined level (conc) for 15 minutes before making an incision. It was then noted whether the patient moved, i.e. jerked or twisted.

Usage

anesthetic

Format

This data frame contains the following columns:

move

a binary numeric vector coded for patient movement (0 = no movement, 1 = movement)

conc

anesthetic concentration

logconc

logarithm of concentration

nomove

the complement of move

Details

The interest is in estimating how the probability of jerking or twisting varies with increasing concentration of the anesthetic agent.

Source

unknown

Examples

print("Logistic Regression - Example 8.1.4")

z <- table(anesthetic$nomove, anesthetic$conc)
tot <- apply(z, 2, sum)         # totals at each concentration
prop <- z[2,  ]/(tot)           # proportions at each concentration
oprop <- sum(z[2,  ])/sum(tot)  # expected proportion moving if concentration had no effect
conc <- as.numeric(dimnames(z)[[2]])
plot(conc, prop, xlab = "Concentration", ylab = "Proportion", xlim = c(.5,2.5),
    ylim = c(0, 1), pch = 16)
chw <- par()$cxy[1]
text(conc - 0.75 * chw, prop, paste(tot), adj = 1)
abline(h = oprop, lty = 2)

pause()

anes.logit <- glm(nomove ~ conc, family = binomial(link = logit),
  data = anesthetic)
anova(anes.logit)
summary(anes.logit)

Description

These data frames have yield averages by blocks (parcels). The ant111bdataset is a subset that has block averages of corn yields for treatment 111 only

Usage

data(antigua)
  data(ant111b)

Format

A data frame with 324 observations on 7 variables.

id

a numeric vector

site

a factor with 8 levels.

block

a factor with levels I II III IV

plot

a numeric vector

trt

a factor consisting of 12 levels

ears

a numeric vector; note that -9999 is used as a missing value code.

harvwt

a numeric vector; the average yield

Source

Andrews DF; Herzberg AM, 1985. Data. A Collection of Problems from Many Fields for the Student and Research Worker. Springer-Verlag. (pp. 339-353)

Description

Each of 20 tasters each assessed three out of the four varieties. The experiment was conducted according to a balanced incomplete block design.

Usage

data(appletaste)

Format

A data frame with 60 observations on the following 3 variables.

aftertaste

a numeric vector

Apple samples were rated for aftertaste, by making a mark on a continuous scale that ranged from 0 (extreme dislike) to 150 (like very much).

panelist

a factor with levels a b c d e f g h i j k l m n o p q r s t

product

a factor with levels 298 493 649 937

Examples

data(appletaste)
appletaste.aov <- aov(aftertaste ~ panelist + product, data=appletaste)
termplot(appletaste.aov)

Description

Distances between the Australian cities of Adelaide, Alice, Brisbane, Broome, Cairns, Canberra, Darwin, Melbourne, Perth and Sydney

Usage

audists

Format

The format is: Class 'dist', i.e., a distance matrix.

Source

Australian road map

Examples

data(audists)
## Not run: 
audists.cmd <- cmdscale(audists)
library(lattice)
xyplot(audists.cmd[,2] ~ audists.cmd[,1], 
       groups=row.names(audists.cmd),
       panel = function(x, y, subscripts, groups)  
                        ltext(x = x, y = y, label = groups[subscripts],
                        cex=1, fontfamily = "HersheySans"))

## End(Not run)

Description

Latitudes and longitudes for Adelaide, Alice, Brisbane, Broome, Cairns, Canberra, Darwin, Melbourne, Perth and Sydney; i.e., for the cities to which the road distances in audists relate.

Usage

aulatlong

Format

A data frame with 10 observations on the following 2 variables.

latitude

Latitude, as a decimal number

longitude

Latitude, as a decimal number

Source

Map of Australia showing latitude and longitude information.

Examples

data(aulatlong)
## maybe str(aulatlong) ; plot(aulatlong) ...

Description

Population figures for Australian states and territories for 1917, 1927, ..., 1997.

Usage

austpop

Format

This data frame contains the following columns:

year

a numeric vector

NSW

New South Wales population counts

Vic

Victoria population counts

Qld

Queensland population counts

SA

South Australia population counts

WA

Western Australia population counts

Tas

Tasmania population counts

NT

Northern Territory population counts

ACT

Australian Capital Territory population counts

Aust

Population counts for the whole country

Source

Australian Bureau of Statistics

Examples

print("Looping - Example 1.7")

growth.rates <- numeric(8)
for (j in seq(2,9)) {
    growth.rates[j-1] <- (austpop[9, j]-austpop[1, j])/austpop[1, j] }
growth.rates <- data.frame(growth.rates)
row.names(growth.rates) <- names(austpop[c(-1,-10)])
  # Note the use of row.names() to name the rows of the data frame
growth.rates

pause()
print("Avoiding Loops - Example 1.7b")

sapply(austpop[,-c(1,10)], function(x){(x[9]-x[1])/x[1]})

pause()
print("Plot - Example 1.8a")
attach(austpop)
plot(year, ACT, type="l") # Join the points ("l" = "line")
detach(austpop)

pause()
print("Exerice 1.12.9")
attach(austpop)
oldpar <- par(mfrow=c(2,4))  
for (i in 2:9){
plot(austpop[,1], log(austpop[, i]), xlab="Year",
    ylab=names(austpop)[i], pch=16, ylim=c(0,10))}
par(oldpar) 
detach(austpop)

Description

Best subset selection applied to completely random noise. This function demonstrates how variable selection techniques in regression can often err in including explanatory variables that are indistinguishable from noise.

Usage

bestsetNoise(m = 100, n = 40, method = "exhaustive", nvmax = 3,
              X = NULL, y=NULL, intercept=TRUE,
              print.summary = TRUE, really.big = FALSE, ...)

bestset.noise(m = 100, n = 40, method = "exhaustive", nvmax = 3,
              X = NULL, y=NULL, intercept=TRUE,
              print.summary = TRUE, really.big = FALSE, ...)

bsnCV(m = 100, n = 40, method = "exhaustive", nvmax = 3,
              X = NULL, y=NULL, intercept=TRUE, nfolds = 2,
              print.summary = TRUE, really.big = FALSE)

bsnOpt(X = matrix(rnorm(25 * 10), ncol = 10), y = NULL, method = "exhaustive",
              nvmax = NULL, nbest = 1, intercept = TRUE, criterion = "cp",
              tcrit = NULL, print.summary = TRUE, really.big = FALSE,
         ...)

bsnVaryNvar(m = 100, nvar = nvmax:50, nvmax = 3, method = "exhaustive",
              intercept=TRUE,
              plotit = TRUE, xlab = "# of variables from which to select",
              ylab = "p-values for t-statistics", main = paste("Select 'best'",
                                                  nvmax, "variables"),
              details = FALSE, really.big = TRUE, smooth = TRUE, ...)

Arguments

Details

If X is not supplied, and in any case for bsnVaryNvar, a set of n predictor variables are simulated as independent standard normal, i.e. N(0,1), variates. Additionally a N(0,1) response variable is simulated. The function bsnOpt selects the ‘best’ model with nvmax or fewer explanatory variables, where the argument criterion specifies the criterion that will be used to choose between models with different numbers of explanatory columns. Other functions select the ‘best’ model with nvmax explanatory columns. In any case, the selection is made using the regsubsets() function from the leaps package. (The leaps package must be installed for this function to work.)

The function bsnCV splits the data (randomly) into nfolds (2 or more) parts. It puts each part aside in turn for use to fit the model (effectively, test data), with the remaining data used for selecting the variables that will be used for fitting. One model fit is returned for each of the nfolds parts.

The function bsnVaryVvar makes repeated calls to bestsetNoise

Value

bestsetNoise returns the lm model object for the "best" model with nvmax explanatory columns.

bsnCV returns as many models as there are folds.

bsnVaryVvar silently returns either (details=FALSE) a matrix that has p-values of the coefficients for the ‘best’ choice of model for each different number of candidate variables, or (details=TRUE) a list with elements:

Matrices have one row for each choice of nvar. The statistics returned are for the ‘best’ model with nvmax explanatory variables.

bsnOpt silently returns a list with elements:

tcritFor each model, the minimum of the absolute values of the t-statistics. regsubsets_objThe object returned by the call to regsubsets.

Note

These functions are primarily designed to demonstrate the biases that can be expected, relative to theoretical estimates of standard errors of parameters and other fitted model statistics, when there is prior selection of the columns that are to be included in the model. With the exception of bsnVaryNvar, they can also be used with an X and y for actual data. In that case, the p-values should be compared with those obtained from repeated use of the function where y is random noise, as a check on the extent of selection effects.

Author(s)

J.H. Maindonald

See Also

lm

Examples

leaps.out <- try(require(leaps, quietly=TRUE))
leaps.out.log <- is.logical(leaps.out)
if ((leaps.out.log==TRUE)&(leaps.out==TRUE)){
bestsetNoise(20,6) # `best' 3-variable regression for 20 simulated observations
                   # on 7 unrelated variables (including the response)
bsnCV(20,6) # `best' 3-variable regressions (one for each fold) for 20
                   # simulated observations on 7 unrelated variables
                   # (including the response)
bsnVaryNvar(m = 50, nvar = 3:6, nvmax = 3, method = "exhaustive",
            plotit=FALSE, details=TRUE)
bsnOpt()
}

Description

The biomass data frame has 135 rows and 8 columns. The rainforest data frame is a subset of this one.

Usage

biomass

Format

This data frame contains the following columns:

dbh

a numeric vector

wood

a numeric vector

bark

a numeric vector

fac26

a factor with 3 levels

root

a numeric vector

rootsk

a numeric vector

branch

a numeric vector

species

a factor with levels Acacia mabellae, C. fraseri, Acmena smithii, B. myrtifolia

Source

J. Ash, Australian National University

References

Ash, J. and Helman, C. (1990) Floristics and vegetation biomass of a forest catchment, Kioloa, south coastal N.S.W. Cunninghamia, 2: 167-182.

Description

Australian regional temperature data, Australian regional rainfall data, and Annual SOI, are given for the years 1900-2021. The regional rainfall and temperature data are area-weighted averages for the respective regions. The Southern Oscillation Index (SOI) is the difference in barometric pressure at sea level between Tahiti and Darwin.

Usage

data("bomregions2021")

Format

These data frames contains the following columns:

Year

Year

seAVt

Southeastern region average temperature (degrees C)

southAVt

Southern temperature

eastAVt

Eastern temperature

northAVt

Northern temperature

swAVt

Southwestern temperature

qldAVt

temperature

nswAVt

temperature

ntAVt

temperature

saAVt

temperature

tasAVt

temperature

vicAVt

temperature

waAVt

temperature

mdbAVt

Murray-Darling basin temperature

ausAVt

Australian average temperature, area-weighted mean

seRain

Southeast Australian annual rainfall (mm)

southRain

Southern rainfall

eastRain

Eastern rainfall

northRain

Northern rainfall

swRain

Southwest rainfall

qldRain

Queensland rainfall

nswRain

NSW rainfall

ntRain

Northern Territory rainfall

saRain

South Australian rainfall

tasRain

Tasmanian rainfall

vicRain

Victorian rainfall

waRain

West Australian rainfall

mdbRain

Murray-Darling basin rainfall

ausRain

Australian average rainfall, area weighted

SOI

Annual average Southern Oscillation Index

sunspot

Yearly mean sunspot number

co2mlo

Moana Loa CO2 concentrations, from 1959

co2law

Moana Loa CO2 concentrations, 1900 to 1978

CO2

CO2 concentrations, composite series

avDMI

Annual average Dipole Mode Index, for the Indian Ocean Dipole, from 1950

Source

Australian Bureau of Meteorology web pages:

Go to the url http://www.bom.gov.au/climate/change/, choose timeseries to display, then click "Download data"

For the SOI data, go to the url http://www.bom.gov.au/climate/enso/.

The CO2 series co2law, for Law Dome ice core data. is from https://data.ess-dive.lbl.gov/portals/CDIAC/.

The Moana Loa CO2 series co2mlo is from Dr. Pieter Tans, NOAA/ESRL (https://gml.noaa.gov/ccgg/trends/)

The series CO2 is a composite series, obtained by adding 0.46 to the Law data for 1900 to 1958, then following this with the Moana Loa data that is avaiable from 1959. The addition of 0.46 brings the average of the Law data into agreement with that for the Moana Loa data for the period 1959 to 1968.

The yearly mean sunspot number is a subset of one of several sunspot series that are available from WDC-SILSO, Royal Observatory of Belgium, Brussels. https://www.sidc.be/silso/datafiles/

The dipole mode index data are from https://ds.data.jma.go.jp/tcc/tcc/products/elnino/index/Readme_iod.txt. Note also https://stateoftheocean.osmc.noaa.gov/sur/ind/dmi.php, which has details of several other such series.

References

D.M. Etheridge, L.P. Steele, R.L. Langenfelds, R.J. Francey, J.-M. Barnola and V.I. Morgan, 1998, Historical CO2 records from the Law Dome DE08, DE08-2, and DSS ice cores, in Trends: A Compendium of Data on Global Change, on line at Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tenn., U.S.A.

Lavery, B., Joung, G. and Nicholls, N. 1997. An extended high-quality historical rainfall dataset for Australia. Australian Meteorological Magazine, 46, 27-38.

Nicholls, N., Lavery, B., Frederiksen, C.\ and Drosdowsky, W. 1996. Recent apparent changes in relationships between the El Nino – southern oscillation and Australian rainfall and temperature. Geophysical Research Letters 23: 3357-3360.

SIDC-team, World Data Center for the Sunspot Index, Royal Observatory of Belgium, Monthly Report on the International Sunspot Number, online catalogue of the sunspot index: https://www.sidc.be/silso/datafiles

Examples

plot(ts(bomregions2021[, c("mdbRain","SOI")], start=1900),
     panel=function(y,...)panel.smooth(bomregions2021$Year, y,...))
avrain <- bomregions2021[,"mdbRain"]
xbomsoi <- with(bomregions2021, data.frame(Year=Year, SOI=SOI,
                cuberootRain=avrain^0.33))
xbomsoi$trendSOI <- lowess(xbomsoi$SOI, f=0.1)$y
xbomsoi$trendRain <- lowess(xbomsoi$cuberootRain, f=0.1)$y
xbomsoi$detrendRain <-
  with(xbomsoi, cuberootRain - trendRain + mean(trendRain))
xbomsoi$detrendSOI <-
  with(xbomsoi, SOI - trendSOI + mean(trendSOI))
## Plot time series avrain and SOI: ts object xbomsoi
plot(ts(xbomsoi[, c("cuberootRain","SOI")], start=1900),
     panel=function(y,...)panel.smooth(xbomsoi$Year, y,...),
     xlab = "Year", main="", ylim=list(c(250, 800),c(-20,25)))
par(mfrow=c(1,2))
rainpos <- pretty(xbomsoi$cuberootRain^3, 6)
plot(cuberootRain ~ SOI, data = xbomsoi,
     ylab = "Rainfall (cube root scale)", yaxt="n")
axis(2, at = rainpos^0.33, labels=paste(rainpos))
mtext(side = 3, line = 0.8, "A", adj = -0.025)
with(xbomsoi, lines(lowess(cuberootRain ~ SOI, f=0.75)))
plot(detrendRain ~ detrendSOI, data = xbomsoi,
     xlab="Detrended SOI", ylab = "Detrended rainfall", yaxt="n")
axis(2, at = rainpos^0.33, labels=paste(rainpos))
with(xbomsoi, lines(lowess(detrendRain ~ detrendSOI, f=0.75)))
mtext(side = 3, line = 0.8, "B", adj = -0.025)
par(mfrow=c(1,1))

Description

The Southern Oscillation Index (SOI) is the difference in barometric pressure at sea level between Tahiti and Darwin. Annual SOI and Australian rainfall data, for the years 1900-2005, are given. Australia's annual mean rainfall is an area-weighted average of the total annual precipitation at approximately 370 rainfall stations around the country.

Usage

bomsoi

Format

This data frame contains the following columns:

Year

a numeric vector

Jan

average January SOI values for each year

Feb

average February SOI values for each year

Mar

average March SOI values for each year

Apr

average April SOI values for each year

May

average May SOI values for each year

Jun

average June SOI values for each year

Jul

average July SOI values for each year

Aug

average August SOI values for each year

Sep

average September SOI values for each year

Oct

average October SOI values for each year

Nov

average November SOI values for each year

Dec

average December SOI values for each year

SOI

a numeric vector consisting of average annual SOI values

avrain

a numeric vector consisting of a weighted average annual rainfall at a large number of Australian sites

NTrain

Northern Territory rain

northRain

north rain

seRain

southeast rain

eastRain

east rain

southRain

south rain

swRain

southwest rain

Source

Australian Bureau of Meteorology web pages:

http://www.bom.gov.au/climate/change/rain02.txt and http://www.bom.gov.au/climate/current/soihtm1.shtml

References

Nicholls, N., Lavery, B., Frederiksen, C.\ and Drosdowsky, W. 1996. Recent apparent changes in relationships between the El Nino – southern oscillation and Australian rainfall and temperature. Geophysical Research Letters 23: 3357-3360.

Examples

plot(ts(bomsoi[, 15:14], start=1900),
     panel=function(y,...)panel.smooth(1900:2005, y,...))
pause()

# Check for skewness by comparing the normal probability plots for 
# different a, e.g.
par(mfrow = c(2,3))
for (a in c(50, 100, 150, 200, 250, 300))
qqnorm(log(bomsoi[, "avrain"] - a))
  # a = 250 leads to a nearly linear plot

pause()

par(mfrow = c(1,1))
plot(bomsoi$SOI, log(bomsoi$avrain - 250), xlab = "SOI",
     ylab = "log(avrain = 250)")
lines(lowess(bomsoi$SOI)$y, lowess(log(bomsoi$avrain - 250))$y, lwd=2)
  # NB: separate lowess fits against time
lines(lowess(bomsoi$SOI, log(bomsoi$avrain - 250)))
pause()

xbomsoi <-
  with(bomsoi, data.frame(SOI=SOI, cuberootRain=avrain^0.33))
xbomsoi$trendSOI <- lowess(xbomsoi$SOI)$y
xbomsoi$trendRain <- lowess(xbomsoi$cuberootRain)$y
rainpos <- pretty(bomsoi$avrain, 5)
with(xbomsoi,
     {plot(cuberootRain ~ SOI, xlab = "SOI",
           ylab = "Rainfall (cube root scale)", yaxt="n")
     axis(2, at = rainpos^0.33, labels=paste(rainpos))
## Relative changes in the two trend curves
     lines(lowess(cuberootRain ~ SOI))
     lines(lowess(trendRain ~ trendSOI), lwd=2)
  })
pause()

xbomsoi$detrendRain <-
  with(xbomsoi, cuberootRain - trendRain + mean(trendRain))
xbomsoi$detrendSOI <-
  with(xbomsoi, SOI - trendSOI + mean(trendSOI))
oldpar <- par(mfrow=c(1,2), pty="s")
plot(cuberootRain ~ SOI, data = xbomsoi,
     ylab = "Rainfall (cube root scale)", yaxt="n")
axis(2, at = rainpos^0.33, labels=paste(rainpos))
with(xbomsoi, lines(lowess(cuberootRain ~ SOI)))
plot(detrendRain ~ detrendSOI, data = xbomsoi,
  xlab="Detrended SOI", ylab = "Detrended rainfall", yaxt="n")
axis(2, at = rainpos^0.33, labels=paste(rainpos))
with(xbomsoi, lines(lowess(detrendRain ~ detrendSOI)))
pause()

par(oldpar)
attach(xbomsoi)
xbomsoi.ma0 <- arima(detrendRain, xreg=detrendSOI, order=c(0,0,0))
# ordinary regression model

xbomsoi.ma12 <- arima(detrendRain, xreg=detrendSOI,
                      order=c(0,0,12))
# regression with MA(12) errors -- all 12 MA parameters are estimated
xbomsoi.ma12
pause()

xbomsoi.ma12s <- arima(detrendRain, xreg=detrendSOI,
                      seasonal=list(order=c(0,0,1), period=12))
# regression with seasonal MA(1) (lag 12) errors -- only 1 MA parameter
# is estimated
xbomsoi.ma12s
pause()

xbomsoi.maSel <- arima(x = detrendRain, order = c(0, 0, 12),
                        xreg = detrendSOI, fixed = c(0, 0, 0,
                        NA, rep(0, 4), NA, 0, NA, NA, NA, NA),
                        transform.pars=FALSE)
# error term is MA(12) with fixed 0's at lags 1, 2, 3, 5, 6, 7, 8, 10
# NA's are used to designate coefficients that still need to be estimated
# transform.pars is set to FALSE, so that MA coefficients are not
# transformed (see help(arima))

detach(xbomsoi)
pause()

Box.test(resid(lm(detrendRain ~ detrendSOI, data = xbomsoi)),
          type="Ljung-Box", lag=20)

pause()

attach(xbomsoi)
 xbomsoi2.maSel <- arima(x = detrendRain, order = c(0, 0, 12),
                         xreg = poly(detrendSOI,2), fixed = c(0,
                         0, 0, NA, rep(0, 4), NA, 0, rep(NA,5)),
                         transform.pars=FALSE)
 xbomsoi2.maSel
qqnorm(resid(xbomsoi.maSel, type="normalized"))
detach(xbomsoi)

Description

The corrected Boston housing data (from http://lib.stat.cmu.edu/datasets/).

Usage

bostonc

Format

A single vector containing the contents of "boston_corrected.txt".

Source

Harrison, D. and Rubinfeld, D.L. 'Hedonic prices and the demand for clean air', J. Environ. Economics & Management, vol.5, 81-102, 1978. corrected by Kelley Pace ([email protected])

Description

Return univariate plotting positions in which neighboring points are separated, if and as necessary, so that they are the specified minimum distance apart.

Usage

bounce(y, d, log = FALSE)

Arguments

Details

The centroid(s) of groups of points that are moved relative to each other remain the same.

Value

A vector of values such that, when plotted along a line, neighboring points are the required minimum distance apart.

Note

If values are plotted on a logarithmic scale, d is the required distance apart on that scale. If a base other than 10 is required, set log equal to that base. (Note that base 10 is the default for plot with log=TRUE.)

Author(s)

John Maindonald

See Also

See also onewayPlot

Examples

bounce(c(4, 1.8, 2, 6), d=.4)
bounce(c(4, 1.8, 2, 6), d=.1, log=TRUE)

Description

This function is useful for use before plotting, if one wants capitalized axis labels or factor levels.

Usage

capstring(names)

Arguments

Value

A character vector with upper case initial values.

Author(s)

W.J. Braun

Examples

capstring(names(tinting)[c(3,4)])

library(lattice)
levels(tinting$agegp) <- capstring(levels(tinting$agegp))
xyplot(csoa ~ it | sex * agegp, data=tinting)

Description

U.S. data extracted from Cars93, a data frame in the MASS package.

Usage

carprice

Format

This data frame contains the following columns:

Type

Type of car, e.g. Sporty, Van, Compact

Min.Price

Price for a basic model

Price

Price for a mid-range model

Max.Price

Price for a ‘premium’ model

Range.Price

Difference between Max.Price and Min.Price

RoughRange

Rough.Range plus some N(0,.0001) noise

gpm100

The number of gallons required to travel 100 miles

MPG.city

Average number of miles per gallon for city driving

MPG.highway

Average number of miles per gallon for highway driving

Source

MASS package

References

Venables, W.N.\ and Ripley, B.D., 4th edn 2002. Modern Applied Statistics with S. Springer, New York.

See also ‘R’ Complements to Modern Applied Statistics with S-Plus, available from http://www.stats.ox.ac.uk/pub/MASS3/

Examples

print("Multicollinearity - Example 6.8")
pairs(carprice[,-c(1,8,9)])

carprice1.lm <- lm(gpm100 ~ Type+Min.Price+Price+Max.Price+Range.Price,
    data=carprice)
round(summary(carprice1.lm)$coef,3)
pause()

alias(carprice1.lm)
pause()

carprice2.lm <- lm(gpm100 ~ Type+Min.Price+Price+Max.Price+RoughRange, data=carprice)
round(summary(carprice2.lm)$coef, 2)
pause()

carprice.lm <- lm(gpm100 ~ Type + Price, data = carprice)
round(summary(carprice.lm)$coef,4)  
pause()

summary(carprice1.lm)$sigma   # residual standard error when fitting all 3 price variables
pause()

summary(carprice.lm)$sigma    # residual standard error when only price is used
pause()

vif(lm(gpm100 ~ Price, data=carprice)) # Baseline Price
pause()

vif(carprice1.lm)    # includes Min.Price, Price & Max.Price
pause()

vif(carprice2.lm)    # includes Min.Price, Price, Max.Price & RoughRange
pause()

vif(carprice.lm)     # Price alone

Description

The Cars93.summary data frame has 6 rows and 4 columns created from information in the Cars93 data set in the Venables and Ripley MASS package. Each row corresponds to a different class of car (e.g. Compact, Large, etc.).

Usage

Cars93.summary

Format

This data frame contains the following columns:

Min.passengers

minimum passenger capacity for each class of car

Max.passengers

maximum passenger capacity for each class of car

No.of.cars

number of cars in each class

abbrev

a factor with levels C Compact, L Large, M Mid-Size, Sm Small, Sp Sporty, V Van

Source

Lock, R. H. (1993) 1993 New Car Data. Journal of Statistics Education 1(1)

References

MASS library

Examples

type <- Cars93.summary$abbrev
type <- Cars93.summary[,4]
type <- Cars93.summary[,"abbrev"]
type <- Cars93.summary[[4]] # Take the object that is stored
                            # in the fourth list element.
type
pause()

attach(Cars93.summary)
  # R can now access the columns of Cars93.summary directly
abbrev
detach("Cars93.summary")
pause()

#  To change the name of the \verb!abbrev! variable (the fourth column)
names(Cars93.summary)[4] <- "code"
pause()

#  To change all of the names, try
names(Cars93.summary) <- c("minpass","maxpass","number","code")

Description

Measurements of sugar content in frosted flakes breakfast cereal.

Usage

cerealsugar

Format

A vector of 100 measurements.

Description

The cfseal data frame has 30 rows and 11 columns consisting of weight measurements for various organs taken from 30 Cape Fur Seals that died as an unintended consequence of commercial fishing.

Usage

cfseal

Format

This data frame contains the following columns:

age

a numeric vector

weight

a numeric vector

heart

a numeric vector

lung

a numeric vector

liver

a numeric vector

spleen

a numeric vector

stomach

a numeric vector

leftkid

a numeric vector

rightkid

a numeric vector

kidney

a numeric vector

intestines

a numeric vector

Source

Stewardson, C.L., Hemsley, S., Meyer, M.A., Canfield, P.J. and Maindonald, J.H. 1999. Gross and microscopic visceral anatomy of the male Cape fur seal, Arctocephalus pusillus pusillus (Pinnepedia: Otariidae), with reference to organ size and growth. Journal of Anatomy (Cambridge) 195: 235-255. (WWF project ZA-348)

Examples

print("Allometric Growth - Example 5.7")

cfseal.lm <- lm(log(heart) ~ log(weight), data=cfseal); summary(cfseal.lm)
plot(log(heart) ~ log(weight), data = cfseal, pch=16, xlab = "Heart Weight (g, log scale)", 
ylab = "Body weight (kg, log scale)", axes=FALSE)
heartaxis <- 100*(2^seq(0,3))
bodyaxis <- c(20,40,60,100,180)
axis(1, at = log(bodyaxis), lab = bodyaxis)
axis(2, at = log(heartaxis), lab = heartaxis)
box()
abline(cfseal.lm)

Description

Population estimates for several Canadian cities.

Usage

cities

Format

This data frame contains the following columns:

CITY

a factor, consisting of the city names

REGION

a factor with 5 levels (ATL=Atlantic, ON=Ontario, QC=Quebec, PR=Prairies, WEST=Alberta and British Columbia) representing the location of the cities

POP1992

a numeric vector giving population in 1000's for 1992

POP1993

a numeric vector giving population in 1000's for 1993

POP1994

a numeric vector giving population in 1000's for 1994

POP1995

a numeric vector giving population in 1000's for 1995

POP1996

a numeric vector giving population in 1000's for 1996

Source

Statistics Canada

Examples

cities$have <- factor((cities$REGION=="ON")|(cities$REGION=="WEST"))
plot(POP1996~POP1992, data=cities, col=as.integer(cities$have))

Description

Data are from trials that studied the mortality response of codling moth to fumigation with methyl bromide.

Usage

data(codling)

Format

A data frame with 99 observations on the following 10 variables.

dose

Injected dose of methyl bromide, in gm per cubic meter

tot

Number of insects in chamber

dead

Number of insects dying

pobs

Proportion dying

cm

Control mortality, i.e., at dose 0

ct

Concentration-time sum

Cultivar

a factor with levels BRAEBURN FUJI GRANNY Gala ROYAL Red Delicious Splendour

gp

a factor which has a different level for each different combination of Cultivar, year and rep (replicate).

year

a factor with levels 1988 1989

numcm

a numeric vector: total number of control insects

Details

The research that generated these data was in part funded by New Zealand pipfruit growers. The published analysis was funded by New Zealand pipfruit growers. See also sorption.

Source

Maindonald, J.H.; Waddell, B.C.; Petry, R.J. 2001. Apple cultivar effects on codling moth (Lepidoptera: Tortricidae) egg mortality following fumigation with methyl bromide. Postharvest Biology and Technology 22: 99-110.

Description

Compare error rates, between different functions and different selection rules, for an approximately equal random division of the data into a training and test set.

Usage

compareTreecalcs(x = yesno ~ ., data = DAAG::spam7, cp = 0.00025, fun = c("rpart",
"randomForest"))

Arguments

Details

Data are randomly divided into two subsets, I and II. The function(s) are used in the standard way for calculations on subset I, and error rates returined that come from the calculations carried out by the function(s). Predictions are made for subset II, allowing the calculation of a completely independent set of error rates.

Value

If rpart is specified in fun, the following:

If rpart is specified in fun, the following:

Author(s)

John Maindonald

Description

Component + Residual plot for a term in a lm model.

Usage

component.residual(lm.obj, which = 1, xlab = "Component",
    ylab = "C+R")

Arguments

Value

A scatterplot with a smooth curve overlaid.

Author(s)

J.H. Maindonald

See Also

lm

Examples

mice12.lm <- lm(brainwt ~ bodywt + lsize, data=litters)
oldpar <- par(mfrow = c(1,2))
component.residual(mice12.lm, 1, xlab = "Body weight", ylab= "t(Body weight) + e")
component.residual(mice12.lm, 2, xlab = "Litter size", ylab= "t(Litter size) + e")
par(oldpar)

Description

Given actual and predicted group assignments, give the confusion matrix

Usage

confusion(actual, predicted, gpnames = NULL, rowcol=c("actual", "predicted"),
printit = c("overall","confusion"), prior = NULL, digits=3)

Arguments

Details

Predicted group assignments should be estimated from cross-validation or from bootstrap out-of-bag data. Better still, work with assignments for test data that are completely separate from the data used to dervive the model.

Value

A list with elements overall (overall accuracy), confusion (confusion matrix) and prior (prior used for calculation of overall accuracy)

Author(s)

John H Maindonald

References

Maindonald and Braun: 'Data Analysis and Graphics Using R', 3rd edition 2010, Section 12.2.2

Examples

library(MASS)
library(DAAG)
cl <- lda(species ~ length+breadth, data=cuckoos, CV=TRUE)$class
confusion(cl, cuckoos$species)

## The function is currently defined as
function (actual, predicted, gpnames = NULL,
            rowcol = c("actual", "predicted"),
            printit = c("overall","confusion"),
            prior = NULL, digits = 3) 
{
  if (is.null(gpnames)) 
    gpnames <- levels(actual)
  if (is.logical(printit)){
    if(printit)printit <- c("overall","confusion")
    else printit <- ""
  }
  tab <- table(actual, predicted)
  acctab <- t(apply(tab, 1, function(x) x/sum(x)))
  dimnames(acctab) <- list(Actual = gpnames, `Predicted (cv)` = gpnames)
  if (is.null(prior)) {
    relnum <- table(actual)
    prior <- relnum/sum(relnum)
    acc <- sum(tab[row(tab) == col(tab)])/sum(tab)
  }
  else {
    acc <- sum(prior * diag(acctab))
  }
  names(prior) <- gpnames
  if ("overall"%in%printit) {
    cat("Overall accuracy =", round(acc, digits), "\n")
    if(is.null(prior)){
      cat("This assumes the following prior frequencies:", 
          "\n")
      print(round(prior, digits))
    }
  }
  if ("confusion"%in%printit) {
    cat("\nConfusion matrix", "\n")
    print(round(acctab, digits))
  }
  invisible(list(overall=acc, confusion=acctab, prior=prior))
}

Description

P-values were calculated for each of 3072 genes, for data that compared expression values between post-settlement coral larvae and pre-settlement coral larvae.

Usage

data("coralPval")

Format

The format is: num [1:3072, 1] 8.60e-01 3.35e-08 3.96e-01 2.79e-01 6.36e-01 ...

Details

t-statistics, and hence p-values, were derived from five replicate two-colour micro-array slides. Details are in a vignette that accompanies the DAAGbio package.

Source

See the ?DAAGbio::coralRG

References

Grasso, L. C.; Maindonald, J.; Rudd, S.; Hayward, D. C.; Saint, R.; Miller, D. J.; and Ball, E. E., 2008. Microarray analysis identifies candidate genes for key roles in coral development. BMC Genomics, 9:540.

Examples

## From p-values, calculate Benjamini-Hochberg false discrimination rates
fdr <- p.adjust(DAAG::coralPval, method='BH')
## Number of genes identified as differentially expressed for FDR = 0.01
sum(fdr<=0.01)

Description

Numbers are given in different categories of worker, in each of two investigations. The first source of information is the Board of Trade Census that was conducted on 1886. The second is a relatively informal survey conducted by US Bureau of Labor representatives in 1889, for use in official reports.

Usage

data(cottonworkers)

Format

A data frame with 14 observations on the following 3 variables.

census1886

Numbers of workers in each of 14 different categories, according to the Board of Trade wage census that was conducted in 1886

survey1889

Numbers of workers in each of 14 different categories, according to data collected in 1889 by the US Bureau of Labor, for use in a report to the US Congress and House of Representatives

avwage

Average wage, in pence, as estimated in the US Bureau of Labor survey

Details

The data in survey1889 were collected in a relatively informal manner, by approaching individuals on the street. Biases might therefore be expected.

Source

United States congress, House of Representatives, Sixth Annual Report of the Commissioner of Labor, 1890, Part III, Cost of Living (Washington D.C. 1891); idem., Seventh Annual Report of the Commissioner of Labor, 1891, Part III, Cost of Living (Washington D.C. 1892)

Return of wages in the principal textile trades of the United Kingdom, with report therein. (P.P. 1889, LXX). United Kingdom Official Publication.

References

Boot, H. M. and Maindonald, J. H. 2007. New estimates of age- and sex- specific earnings and the male-female earnings gap in the British cotton industry, 1833-1906. Economic History Review. Published online 28-Aug-2007 doi: 10.1111/j.1468-0289.2007.00398.x

Examples

data(cottonworkers)
str(cottonworkers)
plot(survey1889 ~ census1886, data=cottonworkers)
plot(I(avwage*survey1889) ~ I(avwage*census1886), data=cottonworkers)

Description

Year and birth, lifespan, etc, of British first class cricketers, born 1840-1960, whose handedness could be determined from information in the Who's who of cricketers. The status (alive=0, dead =1), and lifetime or lifespan, is for 1992.

Usage

data(cricketer)

Format

A data frame with 5960 observations on the following 8 variables.

left

a factor with levels right left

year

numeric, year of birth

life

numeric, lifetime or lifespan to 1992

dead

numeric (0 = alive (censored), 1 = dead, in 1992)

acd

numeric (0 = not accidental or not dead, 1 = accidental death)

kia

numeric (0 = not killed in action, 1 = killed in action)

inbed

numeric (0 = did not die in bed, 1 = died in bed)

cause

a factor with levels alive acd (accidental death) inbed (died in bed)

Details

Note that those 'killed in action' (mostly during World Wars I and II) form a subset of those who died by accident.

Source

John Aggleton, Martin Bland. Data were collated as described in Aggleton et al.

References

Aggleton JP, Bland JM, Kentridge RW, Neave NJ 1994. Handedness and longevity: an archival study of cricketers. British Medical Journal 309, 1681-1684.

Bailey P, Thorne P, Wynne-Thomas P. 1993. Who's Who of Cricketers. 2nd ed, London, Hamlyn.

Bland M and Altman D. 2005. Do the left-handed die young? Significance 2, 166-170.

See Also

earlycrcktr.

Examples

data(cricketer)
numLH <- xtabs(~ left+year, data=cricketer)
propLH <- prop.table(numLH, margin=2)[2,]
yr <- as.numeric(colnames(numLH))
plot(propLH ~ yr)
cricketer$lh <- unclass(cricketer$left)-1
left2.hat <- fitted(lm(lh ~ poly(year,2), data=cricketer))
ord <- order(cricketer$year)
lines(left2.hat[ord] ~ cricketer$year[ord])
library(splines)
ns3.hat <- fitted(lm(lh ~ ns(year,3), data=cricketer))
lines(ns3.hat[ord] ~ cricketer$year[ord], col="red")
require(survival)
summary(coxph(Surv(life, kia) ~ bs(year,3) +left, data=cricketer))
cricketer$notacdDead <- with(cricketer, {dead[acd==1]<-0; dead})
summary(coxph(Surv(life, notacdDead) ~ ns(year,2) +left, data=cricketer))

Description

These data compare mean length, mean breadth, and egg color, between cuckoos and their hosts.

Usage

cuckoohosts

Format

A data frame with 10 observations on the following 12 variables.

clength

mean length of cuckoo eggs in given host's nest

cl.sd

standard deviation of cuckoo egg lengths

cbreadth

mean breadth of cuckoo eggs in given host's nest

cb.sd

standard deviation of cuckoo egg breadths

cnum

number of cuckoo eggs

hlength

length of host eggs

hl.sd

standard deviation of host egg lengths

hbreadth

breadth of host eggs

hb.sd

standard deviation of host egg breadths

hnum

number of host eggs

match

number of eggs where color matched

nomatch

number where color did not match

Details

Although from the same study that generated data in the data frame cuckoos, the data do not match precisely. The cuckoo egg lengths and breadths are from the tables on page 168, the host egg lengths and breadths from Appendix IV on page 176, and the color match counts from the table on page 171.

Source

Latter, O.H., 1902. The egg of cuculus canorus. an inquiry into the dimensions of the cuckoo's egg and the relation of the variations to the size of the eggs of the foster-parent, with notes on coloration, &c. Biometrika, 1:164–176.

Examples

cuckoohosts
str(cuckoohosts)
plot(cuckoohosts)
with(cuckoohosts,
     plot(c(clength,hlength),c(cbreadth,hbreadth),col=rep(1:2,c(6,6))))

Description

Length and breadth measurements of 120 eggs lain in the nests of six different species of host bird.

Usage

cuckoos

Format

This data frame contains the following columns:

length

the egg lengths in millimeters

breadth

the egg breadths in millimeters

species

a factor with levels hedge.sparrow, meadow.pipit, pied.wagtail, robin, tree.pipit, wren

id

a numeric vector

Source

Latter, O.H. (1902). The eggs of Cuculus canorus. An Inquiry into the dimensions of the cuckoo's egg and the relation of the variations to the size of the eggs of the foster-parent, with notes on coloration, &c. Biometrika i, 164. Tippett (1931) gives summary details of the data.

References

Tippett, L.H.C. 1931: "The Methods of Statistics". Williams & Norgate, London.

Examples

print("Strip and Boxplots - Example 2.1.2")

attach(cuckoos)
oldpar <- par(las = 2) # labels at right angle to axis.
stripchart(length ~ species) 
boxplot(split(cuckoos$length, cuckoos$species),
         xlab="Length of egg", horizontal=TRUE)
detach(cuckoos)
par(oldpar)
pause()

print("Summaries - Example 2.2.2")
sapply(split(cuckoos$length, cuckoos$species), sd)
pause()

print("Example 4.1.4")
wren <- split(cuckoos$length, cuckoos$species)$wren
median(wren)
n <- length(wren)
sqrt(pi/2)*sd(wren)/sqrt(n)  # this s.e. computation assumes normality

Description

These functions give training (internal) and cross-validation measures of predictive accuracy for regression with a binary response. The data are randomly divided between a number of ‘folds’. Each fold is removed, in turn, while the remaining data are used to re-fit the regression model and to predict at the omitted observations.

Usage

CVbinary(obj, rand=NULL, nfolds=10, print.details=TRUE)

cv.binary(obj, rand=NULL, nfolds=10, print.details=TRUE)

Arguments

Value

Note

The term ‘training’ seems preferable to the term ‘internal’ in connection with predicted values, and the accuracy measure, that are based on the observations used to derive the model.

Author(s)

J.H. Maindonald

See Also

glm

Examples

frogs.glm <- glm(pres.abs ~ log(distance) + log(NoOfPools),
                 family=binomial,data=frogs)
CVbinary(frogs.glm)
mifem.glm <- glm(outcome ~ ., family=binomial, data=mifem)
CVbinary(mifem.glm)

Description

This function gives internal and cross-validation measures of predictive accuracy for multiple linear regression. (For binary logistic regression, use the CVbinary function.) The data are randomly assigned to a number of ‘folds’. Each fold is removed, in turn, while the remaining data is used to re-fit the regression model and to predict at the deleted observations.

Usage

CVlm(data = DAAG::houseprices, form.lm = formula(sale.price ~ area),
              m = 3, dots = FALSE, seed = 29, plotit = c("Observed","Residual"),
              col.folds=NULL,               
              main="Small symbols show cross-validation predicted values",
              legend.pos="topleft", 
              printit = TRUE, ...)
cv.lm(data = DAAG::houseprices, form.lm = formula(sale.price ~ area),
              m = 3, dots = FALSE, seed = 29, plotit = c("Observed","Residual"),
              col.folds=NULL,               
              main="Small symbols show cross-validation predicted values",
              legend.pos="topleft", printit = TRUE, ...)

Arguments

Details

When plotit="Residual" and there is more than one explanatory variable, the fitted lines that are shown for the individual folds are approximations.

Value

The input data frame is returned, with additional columns Predicted (Predicted values using all observations) and cvpred (cross-validation predictions). The cross-validation residual sum of squares (ss) and degrees of freedom (df) are returned as attributes of the data frame.

Author(s)

J.H. Maindonald

See Also

lm, CVbinary

Examples

CVlm()
## Not run: 
CVlm(data=nihills, form.lm=formula(log(time)~log(climb)+log(dist)),
          plotit="Observed")
CVlm(data=nihills, form.lm=formula(log(time)~log(climb)+log(dist)),
     plotit="Residual")
out <- CVlm(data=nihills, form.lm=formula(log(time)~log(climb)+log(dist)),
               plotit="Observed")
out[c("ms","df")]

## End(Not run)

Description

This generates themes for use in "A Practical Guide to Data Analysis Using R".

Usage

DAAGtheme(fontsize = list(text = 10, points = 6), box = "gray40", color=TRUE,
          sides = list(tck = 0.6, pad1 = 0.75, pad2 = 0.75),...)

Arguments

Details

Setting the color of the bounding box and of the strip boxes to gray, which is the default, reduces the focus on them.

Value

A list which can be used as the par.settings argument to lattice graphics functions, or as the theme argument to trellis.par.set().

Note

The code provides an example of the creation of a functions that generates themes that are tuned to specific user requirements. In this connection, see also theEconomist.theme.

Author(s)

John Maindonald.

See Also

standard.theme, simpleTheme, theEconomist.theme, custom.theme

Examples

bwtheme <- DAAGtheme(pch=2:4)
lattice::xyplot(csoa ~ age | target, groups=sex, 
                data=DAAG::tinting, par.settings=bwtheme)

Description

This is the default database for use with the function datafile, which uses elements of this list to place files in the working directory.

Usage

data(DAAGxdb)

Format

Successive elements in this list hold character vectors from which the corresponding files can be generated. The names of the list elements are fuel, fuel.csv, oneBadRow, scan-demo, molclock1, molclock2, and travelbooks.

Details

The files fuel.txt and fuel.csv are used in Chapter 1 of DAAGUR, while the files oneBadRow.txt and scan-demo.txt are used in Chapter 14 of DAAGUR.

References

Maindonald, J.H. and Braun, W.J. 2007. Data Analysis and Graphics Using R: An Example-Based Approach. 2nd edn, Cambridge University Press (DAAGUR).

Examples

data(DAAGxdb)
names(DAAGxdb)

Description

Invoking this function writes one or more nominated files to the working directory. In particular, it may be used to write the files 'fuel.txt' and 'fuel.csv' that are used in Chapter 1 of DAAGUR, and the files 'oneBadRow.txt' and 'scan-demo.txt' that are used in Chapter 14 of DAAGUR.

Usage

datafile(file = c("fuel", "travelbooks"), datastore =
                 DAAG::DAAGxdb, altstore = DAAG::zzDAAGxdb, showNames =
                 FALSE)

Arguments

Value

An ASCII file is output to the current working directory. The names of all available datasets are returned invisibly.

Author(s)

J.H. Maindonald

Examples

datafile(file="", showNames=TRUE)

Description

Data record, for each of 2000 administrative regions, whether or not dengue was recorded at any time between 1961 and 1990.

Usage

data(dengue)

Format

A data frame with 2000 observations on the following 13 variables.

humid

Average vapour density: 1961-1990

humid90

90th percentile of humid

temp

Average temperature: 1961-1990

temp90

90th percentile of temp

h10pix

maximum of humid, within a 10 pixel radius

h10pix90

maximum of humid90, within a 10 pixel radius

trees

Percent tree cover, from satellite data

trees90

90th percentile of trees

NoYes

Was dengue observed? (1=yes)

Xmin

minimum longitude

Xmax

maximum longitude

Ymin

minimum latitude

Ymax

maximum latitude

Details

This is derived from a data set in which the climate and tree cover information were given for each half degree of latitude by half degreee of longitude pixel. The variable NoYes was given by administrative region. The climate data and tree cover data given here are 50th or 90th percentiles, where percetiles were calculates across pixels for an administrative region.

Source

Simon Hales, Environmental Research New Zealand Ltd.

References

Hales, S., de Wet, N., Maindonald, J. and Woodward, A. 2002. Potential effect of population and climate change global distribution of dengue fever: an empirical model. The Lancet 2002; 360: 830-34.

Examples

str(dengue)
glm(NoYes ~ humid, data=dengue, family=binomial)
glm(NoYes ~ humid90, data=dengue, family=binomial)

Description

The dewpoint data frame has 72 rows and 3 columns. Monthly data were obtained for a number of sites (in Australia) and a number of months.

Usage

dewpoint

Format

This data frame contains the following columns:

maxtemp

monthly minimum temperatures

mintemp

monthly maximum temperatures

dewpt

monthly average dewpoint for each combination of minimum and maximum temperature readings (formerly dewpoint)

Source

Dr Edward Linacre, visiting fellow in the Australian National University Department of Geography.

Examples

print("Additive Model - Example 7.5")
require(splines)
attach(dewpoint)   
ds.lm <- lm(dewpt ~ bs(maxtemp,5) + bs(mintemp,5), data=dewpoint)
ds.fit <-predict(ds.lm, type="terms", se=TRUE)
oldpar <- par(mfrow=c(1,2))
plot(maxtemp, ds.fit$fit[,1], xlab="Maximum temperature",
     ylab="Change from dewpoint mean",type="n")
lines(maxtemp,ds.fit$fit[,1])
lines(maxtemp,ds.fit$fit[,1]-2*ds.fit$se[,1],lty=2)
lines(maxtemp,ds.fit$fit[,1]+2*ds.fit$se[,1],lty=2)
plot(mintemp,ds.fit$fit[,2],xlab="Minimum temperature",
     ylab="Change from dewpoint mean",type="n")
ord<-order(mintemp)
lines(mintemp[ord],ds.fit$fit[ord,2])
lines(mintemp[ord],ds.fit$fit[ord,2]-2*ds.fit$se[ord,2],lty=2)
lines(mintemp[ord],ds.fit$fit[ord,2]+2*ds.fit$se[ord,2],lty=2)
detach(dewpoint)
par(oldpar)

Description

Data collected at Winnipeg International Airport (Canada) on periods (in days) between rain events.

Usage

droughts

Format

This data frame contains the following columns:

length

the length of time from the completion of the last rain event to the beginning of the next rain event.

year

the calendar year.

Examples

boxplot(length ~ year, data=droughts)
  boxplot(log(length) ~ year, data=droughts)
  hist(droughts$length, main="Winnipeg Droughts", xlab="length (in days)")
  hist(log(droughts$length), main="Winnipeg Droughts", xlab="length (in days, log scale)")

Description

Carbon dioxide record from the EPICA (European Project for Ice Coring in Antarctica) Dome C ice core covering 0 to 800 kyr BP.

Usage

data(edcCO2)

Format

A data frame with 1096 observations on the following 2 variables.

age

Age in years before present (BP)

co2

CO2 level (ppmv)

Details

Data are a composite series.

Source

Go to the url https://www.ncei.noaa.gov/products/paleoclimatology/ice-core/

References

Luthi, D., M. et al. 2008. High-resolution carbon dioxide concentration record 650,000-800,000 years before present. Nature, Vol. 453, pp. 379-382, 15 May 2008. doi:10.1038/nature06949

Indermuhle, A., E. et al, 1999, Atmospheric CO2 concentration from 60 to 20 kyr BP from the Taylor Dome ice core, Antarctica. Geophysical Research Letters, 27, 735-738.

Monnin, E., A. et al. 2001. Atmospheric CO2 concentrations over the last glacial termination. Science, Vol. 291, pp. 112-114.

Petit, J.R. et al. 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature 399: 429-436.

Siegenthaler, U. et al. 2005. Stable Carbon Cycle-Climate Relationship During the Late Pleistocene. Science, v. 310 , pp. 1313-1317, 25 November 2005.

Examples

data(edcCO2)

Description

Temperature record, using Deuterium as a proxy, from the EPICA (European Project for Ice Coring in Antarctica) Dome C ice core covering 0 to 800 kyr BP.

Usage

data(edcT)

Format

A data frame with 5788 observations on the following 5 variables.

Bag

Bag number

ztop

Top depth (m)

Age

Years before 1950

Deuterium

Deuterium dD data

dT

Temperature difference from the average of the last 1000 years ~ -54.5degC

Details

Temperature was estimated from the deuterium data, after making various corrections.

Source

Go to the url https://www.ncei.noaa.gov/products/paleoclimatology/ice-core/

References

Jouzel, J., et al. 2007. EPICA Dome C Ice Core 800KYr Deuterium Data and Temperature Estimates. IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series # 2007-091. NOAA/NCDC Paleoclimatology Program, Boulder CO, USA.

Jouzel, J., et al. 2007. Orbital and Millennial Antarctic Climate Variability over the Past 800,000 Years. Science, Vol. 317, No. 5839, pp.793-797, 10 August 2007.

Examples

data(edcT)

Description

Both datasets give, for each amount by which an elastic band is stretched over the end of a ruler, the distance that the band traveled when released. The elastic1 data frame has 7 rows. The elastic2 data frame, whose data span a wider range of stretches and distances, has 9 rows.

Usage

data(elastic1)
  data(elastic2)

Format

These data frames contains the following columns:

stretch

the amount by which the elastic band was stretched

distance

the distance traveled

Source

J. H. Maindonald

Examples

plot(elastic1)
sapply(elastic1, mean)
pause()
sapply(elastic1, function(x)mean(x))
pause()
sapply(elastic1, function(x)sum(log(x)))
pause()
yrange <- range(c(elastic1$distance, elastic2$distance))
xrange <- range(c(elastic1$stretch, elastic2$stretch))
plot(distance ~ stretch, data = elastic1, pch = 16, ylim = yrange, xlim = 
xrange)
points(distance ~ stretch, data = elastic2, pch = 15, col = 2)
legend(xrange[1], yrange[2], legend = c("Data set 1", "Data set 2"), pch = 
c(16, 15), col = c(1, 2))
elastic1.lm <- lm(distance ~ stretch, data = elastic1)
elastic2.lm <- lm(distance ~ stretch, data = elastic2)
abline(elastic1.lm)
abline(elastic2.lm, col = 2)
summary(elastic1.lm)
summary(elastic2.lm)
pause()
predict(elastic1.lm, se.fit=TRUE)
predict(elastic2.lm, se.fit=TRUE)

Description

The elasticband data frame has 7 rows and 2 columns giving, for each amount by which an elastic band is stretched over the end of a ruler, the distance that the band traveled when released.

Usage

elasticband

Format

This data frame contains the following columns:

stretch

the amount by which the elastic band was stretched

distance

the distance traveled

Source

J. H. Maindonald

Examples

print("Example 1.8.1")

attach(elasticband)     # R now knows where to find stretch and distance
plot(stretch, distance) # Alternative: plot(distance ~ stretch)
detach(elasticband)

print("Lists - Example 12.7")

elastic.lm <- lm(distance ~ stretch, data=elasticband)
 names(elastic.lm)
 elastic.lm$coefficients
elastic.lm[["coefficients"]]
pause()

elastic.lm[[1]]
pause()

elastic.lm[1]
pause()

options(digits=3)
elastic.lm$residuals 
pause()

elastic.lm$call
pause()

 mode(elastic.lm$call)

Description

Simulates $y-$ and $x-$values for a classical “errors in $x$” linear regression model. One or more $x-$values are subject to random measurement error, independently of the corresponding covariate values that are measured without error.

Usage

errorsINseveral(n = 1000, a0 = 2.5, beta = c(1.5, 0), mu = 12.5, SDyerr = 0.5,
default.Vpar = list(SDx = 2, rho = -0.5, timesSDx = 1.5),
V = with(default.Vpar, matrix(c(1, rho, rho, 1), ncol = 2) * SDx^2),
xerrV = with(default.Vpar, matrix(c(1, 0, 0, 0), ncol = 2) * (SDx * timesSDx)^2),
parset = NULL, print.summary = TRUE, plotit = TRUE)

Arguments

Details

With default arguments, simulates a model in which two covariates are in contention, the first measured without error, and the second with coefficient 0 in the model that includes both covariates measured without error.

Value

Author(s)

John Maindonald

References

Data Analysis and Graphics Using R, 3rd edn, Section 6.8.1

See Also

errorsINx

Examples

library(lattice)
function(n=1000, a0=2.5, beta=c(1.5,0), mu=12.5, SDyerr=0.5, 
           default.Vpar=list(SDx=2, rho=-0.5, timesSDx=1.5),
           V=with(default.Vpar, matrix(c(1,rho,rho,1), ncol=2)*SDx^2),
           xerrV=with(default.Vpar, matrix(c(1,0,0,0), ncol=2)*(SDx*timesSDx)^2),
           parset=NULL, print.summary=TRUE, plotit=TRUE){
    m <- dim(V)[1]
    if(length(mu)==1)mu <- rep(mu,m)
    ow <- options(warn=-1)
    xxmat <- sweep(matrix(rnorm(m*n, 0, 1), ncol=m) %*% chol(V), 2, mu, "+")
    errxx <- matrix(rnorm(m*n, 0, 1), ncol=m) %*% chol(xerrV, pivot=TRUE)
    options(ow)
    dimnames(xxmat)[[2]] <- paste("z", 1:m, sep="")
    xxWITHerr <- xxmat+errxx
    xxWITHerr <- data.frame(xxWITHerr)
    names(xxWITHerr) <- paste("xWITHerr", 1:m, sep="")
    xxWITHerr[, "y"] <- a0 + xxmat %*% matrix(beta,ncol=1) + rnorm(n, sd=SDyerr)
    err.lm <- lm(y ~ ., data=xxWITHerr)
    xx <- data.frame(xxmat)
    names(xx) <- paste("z", 1:m, sep="")
    xx$y <- xxWITHerr$y
    xx.lm <- lm(y ~ ., data=xx)
    B <- coef(err.lm)
    b <- coef(xx.lm)
    SE <- summary(err.lm)$coef[,2]
    se <- summary(xx.lm)$coef[,2]
    if(print.summary){
      beta0 <- c(mean(xx$y)-sum(beta*apply(xx[,1:m],2,mean)), beta)
      tab <- rbind(beta0, b, B)
      dimnames(tab) <- list(c("Values for simulation",
                              "Estimates: no error in x1",
                              "LS Estimates: error in x1"),
                            c("Intercept", paste("b", 1:m, sep="")))
      tabSE <- rbind(rep(NA,m+1),se,SE)
      rownames(tabSE) <- rownames(tab)
      colnames(tabSE) <- c("SE(Int)", paste("SE(", colnames(tab)[-1],")", sep=""))
      tab <- cbind(tab,tabSE)
      print(round(tab,3))
    }
    if(m==2 & print.summary){
      tau <- default.Vpar$timesSDx
      s1 <- sqrt(V[1,1])
      s2 <- sqrt(V[2,2])
      rho <- default.Vpar$rho
      s12 <- s1*sqrt(1-rho^2)
      lambda <- (1-rho^2)/(1-rho^2+tau^2)
      gam12 <- rho*sqrt(V[1,1]/V[2,2])
      expB2 <- beta[2]+beta[1]*(1-lambda)*gam12
      print(c("Theoretical attenuation of b1" = lambda, "Theoretical b2" = expB2))
    }
    if(is.null(parset))parset <- simpleTheme(col=c("gray40","gray40"),
                                             col.line=c("black","black"))
    if(plotit){
      library(lattice)
      zhat <- fitted(xx.lm)
      xhat <- fitted(err.lm)
      plt <- xyplot(xhat ~ zhat, aspect=1, scales=list(tck=0.5),
                    panel=function(x,y,...){
                      panel.xyplot(x,y,type="p",...)
                      panel.abline(lm(y ~ x), lty=2)
                      panel.abline(0,1)
                    },
                    xlab="Fitted values; regress on exact z",
                    ylab="Fitted values; regress on x = xWITHerr",
                    key=list(space="top", columns=2,
                      text=list(lab=c("Line y=x", "Regression fit to points")),
                      lines=list(lty=1:2)),
                    par.settings=parset
                    )
      print(plt)}
    invisible(list(ERRfree=xx, addedERR=xxWITHerr))
  }

Description

Simulates $y-$ and $x-$values for the straight line regression model, but with $x-$values subject to random measurement error, following the classical “errors in x” model. Optionally, the x-values can be split into two groups, with one group shifted relative to the other

Usage

errorsINx(mu = 12.5, n = 200, a = 15, b = 1.5, SDx=2, SDyerr = 1.5,
           timesSDx=(1:5)/2.5, gpfactor=if(missing(gpdiff))FALSE else TRUE,
           gpdiff=if(gpfactor) 1.5 else 0, layout=NULL,
           parset = simpleTheme(alpha = 0.75, col = c("black","gray45"),
             col.line = c("black","gray45"), lwd=c(1,1.5), pch=c(1,2),
           lty=c(1,2)), print.summary=TRUE, plotit=TRUE, xrelation="same")

Arguments

Details

The argument timesSDx can be a numeric vector. One set of $x$-values that are contaminated with measurement error is simulated for each element of timesSDx.

Value

Author(s)

John Maindonald

References

Data Analysis and Graphics Using R, 3rd edn, Section 6.7

Examples

library(lattice)
errorsINx()
errorsINx(gpdiff=2, timesSDx=1.25, SDyerr=2.5, n=80)

Description

This function creates a multi-way table of counts for the response given a set of classifying factors. Output facilitates a check on how the factor specified as margin may, after accounting for other classifying factors, affect the response.

Usage

excessRisk(form = weight ~ seatbelt + airbag, response = "dead", margin = "airbag",
data = DAAG::nassCDS, decpl = 4, printResults = TRUE)

Arguments

Details

The best way to understand what this function does may be to run it with the default parameters, and/or with examples that appear below.

Value

The function returns a data frame, with one row for each combination of levels of factors on the right of the formula, but excluding the factor specified as margin. The final three columns show the count for level 1 as a fraction of the margin by total, the count for level 2 as a fraction of the margin by total, and the excess count for level 2 of response in the row, under the assumption that, in that row, there is no association between response and margin. This is the observed response (for the default arguments, number of dead) for level 2 (airbag deployed), less the number that would have been expected if the proportion had been that for level 1. (Negative values favor airbags.)

Author(s)

John Maindonald

References

See help(nassCDS)

See Also

xtabs

Examples

excessRisk()
excessRisk(weight ~ airbag+seatbelt+dvcat)
UCB <- as.data.frame.table(UCBAdmissions)
excessRisk(Freq~Gender, response="Admit", margin="Gender",data=UCB)
excessRisk(Freq~Gender+Dept, response="Admit", margin="Gender",data=UCB)

Description

Estimates of total worldwide carbon emissions from fossil fuel use.

Usage

fossilfuel

Format

This data frame contains the following columns:

year

a numeric vector giving the year the measurement was taken.

carbon

a numeric vector giving the total worldwide carbon emissions from fossil fuel use, in millions of tonnes.

Details

Data for the years 1751 through to 2014 is available from Data for the years 2014 https://data.ess-dive.lbl.gov/portals/CDIAC

Source

Boden T A; Marland G; Andres R J (1999): Global, Regional, and National Fossil-Fuel CO2 Emissions (1751 - 2014) (V. 2017). Carbon Dioxide Information Analysis Center (CDIAC), Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States), ESS-DIVE repository. Dataset. doi:10.3334/CDIAC/00001_V2017

Examples

plot(fossilfuel)

Description

The frogs data frame has 212 rows and 11 columns. The data are on the distribution of the Southern Corroboree frog, which occurs in the Snowy Mountains area of New South Wales, Australia.

Usage

frogs

Format

This data frame contains the following columns:

pres.abs

0 = frogs were absent, 1 = frogs were present

northing

reference point

easting

reference point

altitude

altitude , in meters

distance

distance in meters to nearest extant population

NoOfPools

number of potential breeding pools

NoOfSites

(number of potential breeding sites within a 2 km radius

avrain

mean rainfall for Spring period

meanmin

mean minimum Spring temperature

meanmax

mean maximum Spring temperature

Source

Hunter, D. (2000) The conservation and demography of the southern corroboree frog (Pseudophryne corroboree). M.Sc. thesis, University of Canberra, Canberra.

Examples

print("Multiple Logistic Regression - Example 8.2")

plot(northing ~ easting, data=frogs, pch=c(1,16)[frogs$pres.abs+1],
  xlab="Meters east of reference point", ylab="Meters north")
pairs(frogs[,4:10])
attach(frogs)
pairs(cbind(altitude,log(distance),log(NoOfPools),NoOfSites),
  panel=panel.smooth, labels=c("altitude","log(distance)",
  "log(NoOfPools)","NoOfSites"))
detach(frogs)

frogs.glm0 <- glm(formula = pres.abs ~ altitude + log(distance) +
  log(NoOfPools) + NoOfSites + avrain + meanmin + meanmax,
  family = binomial, data = frogs)
summary(frogs.glm0)

frogs.glm <- glm(formula = pres.abs ~ log(distance) + log(NoOfPools) + 
meanmin +
  meanmax, family = binomial, data = frogs)
oldpar <- par(mfrow=c(2,2))
termplot(frogs.glm, data=frogs)

termplot(frogs.glm, data=frogs, partial.resid=TRUE)

cv.binary(frogs.glm0)   # All explanatory variables
pause()

cv.binary(frogs.glm)    # Reduced set of explanatory variables

for (j in 1:4){
 rand <- sample(1:10, 212, replace=TRUE)
 all.acc <- cv.binary(frogs.glm0, rand=rand, print.details=FALSE)$acc.cv
 reduced.acc <- cv.binary(frogs.glm, rand=rand, print.details=FALSE)$acc.cv
 cat("\nAll:", round(all.acc,3), "  Reduced:", round(reduced.acc,3))
}

Description

The frosted flakes data frame has 100 rows and 2 columns giving the sugar concentration (in percent) for 25 g samples of a cereal as measured by 2 methods – high performance liquid chromatography (a slow accurate lab method) and a quick method using the infra-analyzer 400.

Usage

frostedflakes

Format

This data frame contains the following columns:

Lab

careful laboratory analysis measurements using high performance liquid chromatography

IA400

measurements based on the infra-analyzer 400

Source

W. J. Braun

Description

Data are from a study that examined how the electrical resistance of a slab of kiwifruit changed with the apparent juice content.

Usage

fruitohms

Format

This data frame contains the following columns:

juice

apparent juice content (percent)

ohms

electrical resistance (in ohms)

Source

Harker, F. R. and Maindonald J.H. 1994. Ripening of nectarine fruit. Plant Physiology 106: 165 - 171.

Examples

plot(ohms ~ juice, xlab="Apparent juice content (%)",ylab="Resistance (ohms)", data=fruitohms)
lines(lowess(fruitohms$juice, fruitohms$ohms), lwd=2)
pause()

require(splines)
attach(fruitohms)
plot(ohms ~ juice, cex=0.8, xlab="Apparent juice content (%)",
     ylab="Resistance (ohms)", type="n")
fruit.lmb4 <- lm(ohms ~ bs(juice,4))
ord <- order(juice)
lines(juice[ord], fitted(fruit.lmb4)[ord], lwd=2)
ci <- predict(fruit.lmb4, interval="confidence")
lines(juice[ord], ci[ord,"lwr"])
lines(juice[ord], ci[ord,"upr"])

Description

The table shows, separately for males and females, the effect of pentazocine on post-operative pain profiles (average VAS scores), with (mbac and fbac) and without (mpl and fpl) preoperatively administered baclofen. Pain scores are recorded every 20 minutes, from 10 minutes to 170 minutes.

Usage

gaba

Format

A data frame with 9 observations on the following 7 variables.

min

a numeric vector

mbac

a numeric vector

mpl

a numeric vector

fbac

a numeric vector

fpl

a numeric vector

avbac

a numeric vector

avplac

a numeric vector

Details

15 females were given baclofen, as against 3 males. 7 females received the placebo, as against 16 males. Averages for the two treatments (baclofen/placebo), taken over all trial participants and ignoring sex, are misleading.

Source

Gordon, N. C. et al.(1995): 'Enhancement of Morphine Analgesia by the GABA$_B$ against Baclofen'. Neuroscience 69: 345-349.

Examples

data(gaba)
mr <- range(gaba$min)
tran <- range(gaba[, c("mbac","mpl","fbac","fpl")])
## Means by treatment and sex
par(mfrow=c(1,2))
plot(mr, tran, xlab = "Time post pentazocine (min)",
     ylab = "Reduction in VAS pain rating", 
     type = "n", xlim = c(0, 170), ylim = tran)
points(gaba$min, gaba$fbac, pch = 1, col = 8, lwd = 2, lty = 2, 
       type = "b")
points(gaba$min, gaba$fpl, pch = 0, col = 8, lwd = 2, lty = 2, 
       type = "b")
points(gaba$min, gaba$mbac, pch = 16, col = 8, lty = 2, type = "b")
points(gaba$min, gaba$mpl, pch = 15, col = 8, lty = 2, type = "b")
box()
## Now plot means, by treatment, averaged over all participants
plot(mr, tran, xlab = "Time post pentazocine (min)",
     ylab = "Reduction in VAS pain rating", 
     type = "n", xlim = c(0, 170), ylim = tran)
bac <- (15 * gaba$fbac + 3 * gaba$mbac)/18
plac <- (7 * gaba$fpl + 9 * gaba$mpl)/16
points(gaba$min, plac, pch = 15, lty = 1, col=1, type = "b")
points(gaba$min, bac, pch = 16, lty = 1, col=1, type = "b")
box()
par(mfrow=c(1,1))

Description

The geophones data frame has 56 rows and 2 columns. Thickness of a layer of Alberta substratum as measured by a line of geophones.

Usage

geophones

Format

This data frame contains the following columns:

distance

location of geophone.

thickness

time for signal to pass through substratum.

Examples

plot(geophones)
lines(lowess(geophones, f=.25))

Description

Heights, stored as a multivariate time series, are for the lakes Erie, Michigan/Huron, Ontario and St Clair

Usage

data(greatLakes)

Format

The format is: mts [1:92, 1:4] 174 174 174 174 174 ... - attr(*, "dimnames")=List of 2 ..$ : NULL ..$ : chr [1:4] "Erie" "michHuron" "Ontario" "StClair" - attr(*, "tsp")= num [1:3] 1918 2009 1 - attr(*, "class")= chr [1:2] "mts" "ts"

Details

For more details, go to the website that is the source of the data.

Source

https://www.lre.usace.army.mil/Missions/Great-Lakes-Information/Great-Lakes-Information-2/Water-Level-Data/

Examples

data(greatLakes)
plot(greatLakes)
## maybe str(greatLakes)

Description

Data are annual apparent alcohol consumption in Australia and New Zealand, in liters of pure alcohol content per annum, separately for beer, wine, and spirits (including spirit-based products).

Usage

data(grog)

Format

A data frame with 18 observations on the following 5 variables.

Beer

liters per annum

Wine

liters per annum

Spirit

liters per annum

Country

a factor with levels Australia NewZealand

Year

Year ending in June of the given year

Details

Data are total available pure alcohol content, for the three categories, divided by numbers of persons aged 15 years or more. The source data for New Zealand included quarterly figures from December 1997, and annual data to December for all years. The annual New Zealand figure to June 1998 required an estimate for September 1997 that was obtained by extrapolating back the third quarter trend line from later years.

Source

Australian data are from https://www.abs.gov.au. For the most recent data, go to https://www.abs.gov.au/statistics/health/health-conditions-and-risks/apparent-consumption-alcohol-australia For New Zealand data, go to https://infoshare.stats.govt.nz/ Click on 'Industry sectors' and then on 'Alcohol Available for Consumption - ALC'.

Examples

data(grog)
library(lattice)
xyplot(Beer+Wine+Spirit ~ Year | Country, data=grog)
xyplot(Beer+Wine+Spirit ~ Year, groups=Country, data=grog, outer=TRUE)