taojiea1014的博客

The response variable(a) Continuous Normal regression, ANOVA or ANCOVA(b) Proportion Logistic regression(c) Count Log-linear models(d) Binary Binary logistic analysis(e) Time at death Survival analysi...

 set the problem of drawing lines to join the nearest neighbour pairs of any given  set of points (x, y) that are mapped in two dimensions. There are three steps to the computing: compute the distance...

, you know the identity of each individual (unlike cluster analysis) and you want to know how theexplanatory variables contribute to the correct classification of individuals. The method works by unco...

rnos <- runif(100)which(rnos<= 0.1)which(rnos<= 0.1)[1]death1 <- numeric(30)for (i in 1:30){rnos <- runif(100)death1[i] <- which(rnos<= 0.1)[1]}death11/mean(death1)death2 <- nu...

For random effects we are often more interested in the question of how much of the variation in the responsevariable can be attributed to a given factor, than we are in estimating means or assessing t...

11.3.1 Split-plot experimentsIn a split-plot experiment, different treatments are applied to plots of different sizes. Each different plot size is associated with its own error variance, so instead of...

 analysis of covariance:some experiments combine regression and ANOVA by fitting a series  of regression lines, one in each of several levels of a given factor .One-way ANOVA: to compare several means...

A factorial experiment has two or more factors, each with two or more levels, plus replication for each combination of factors levels.        This means that we can investigate statistical interaction...

8.1.1 Data summarydata <- read.table("c:\\temp\\classic.txt",header=T)names(data)attach(data)par(mfrow=c(2,2))plot(y)boxplot(y)hist(y,main="")y2 <- yy2[52] <- 21.75plot(y2)summary(y)fivenum(y...

regrowth <- read.table("c:\\temp\\ipomopsis.txt",header=T)attach(regrowth)names(regrowth)plot(Root,Fruit,pch=16,col=c("blue","red")[as.numeric(Grazing)])levels(Grazing)abline(lm(Fruit[Grazing=="Gra...

Gain <- read.table("c:\\temp\\Gain.txt",header=T)attach(Gain)names(Gain)m1 <- lm(Weight~Sex*Age*Genotype)summary(m1)m2 <- step(m1)summary(m2)newGenotype <- Genotypelevels(newGenotype)level...

Ancovacontrasts <- read.table("c:\\temp\\Ancovacontrasts.txt",header=T)attach(Ancovacontrasts)names(Ancovacontrasts)lm(weight[sex=="male"]~age[sex=="male"])lm(weight~age,subset=(sex=="female"))lm(w...

summary.aov(lm(weight~sex*age))summary.aov(lm(weight~age*sex))summary.aov(lm(Fruit~Grazing*Root))summary.aov(lm(Fruit~Root*Grazing))tapply(Root,Grazing, mean)summary(lm(Fruit~Root*Grazing))summary(lm(...

8.4.1 Comparing two variancesBefore we can carry out a test to compare two sample means (see below), we need to test whether the samplevariances are significantly different (see p. 356). The test coul...

There are three broad classes of spatial pattern on a continuum from complete regularity (evenly spaced hexagons where every individual is the same distance from its nearest neighbour) to complete agg...

kmd <- read.table("c:\\temp\\kmeansdata.txt",header=T)attach(kmd)names(kmd)par(mfrow=c(2,2))plot(x,y,pch=16)plot(x,y,col=group,pch=16)model <- kmeans(data.frame(x,y),6)plot(x,y,col=model[[1]])mo...

An alternative to estimating confidence intervals on the regression parameters from the pooled error variancein the ANOVA table (p. 459) is to use bootstrapping. There are two ways of doing this: samp...

A second alternative to estimating confidence intervals on regression parameters is to jackknife the data.Each point in the data set is left out, one at a time, and the parameter of interest is re-est...

The Durbin–Watson function is used for testing whether there is autocorrelation in the residuals froma linear model or a generalized linear model, and is implemented as part of the car package (seeFox...

his kind of regression fits different functions over different ranges of the explanatory variable. For example,it might fit different linear regressions to the left- and right-hand halves of a scatter...

It was Eisenhart (1947) who realized that there were actually two fundamentally different sorts of categoricalexplanatory variables: he called these fixed effects and random effects. It will take a go...

Hierarchical data are often encountered in observational studies where information is collected at a range ofdifferent spatial scales. The principal aim is to discover the scale at which most of the v...

install.packages("tree")library(tree)Pollute <- read.table("c:\\temp\\Pollute.txt",header=T)attach(Pollute)names(Pollute)model <- tree(Pollute)plot(model)text(model)The model is fitted using bin...

epilobium <- read.table("c:\\temp\\epilobium.txt",header=T)attach(epilobium)epilobiummodel <- tree(species ~ .,epilobium,mindev=1e-6,minsize=2)plot(model)text(model,cex=0.7)taxonomy <- read.t...

car.test.frame <- read.table("c:\\temp\\car.test.frame.txt",header=T)attach(car.test.frame)names(car.test.frame)plot(Weight,Mileage,pch=21,col="brown",bg="green")a <- mean(Mileage[Weight<2567...

The three key concepts in time series analysis are trend, serial dependence, and stationarityblowfly <- read.table("c:\\temp\\blowfly.txt",header=T)attach(blowfly)names(blowfly)flies <- ts(flies...

It is useful to be able to turn a time series into components. The function stl (with a lower-case letter L, notnumeral one) performs seasonal decomposition of a time series into seasonal, trend and i...

numbers <- read.table("c:\\temp\\lynx.txt",header=T)attach(numbers)names(numbers)plot.ts(Lynx)The fundamental tool of spectral analysis is the periodogram. This is based on the squared correlationb...

Typically, we would use the partial autocorrelation plot (above) to determine the order. So, for the lynx data(p. 800) we would use order 2 or 4, depending on taste. Other things being equal, parsimon...

streams <- read.table("c:\\temp\\streams.txt",header=T)attach(streams)names(streams)t.test(down,up)t.test(down,up,paired=TRUE)difference <- up - downt.test(difference)8.6 The sign testbinom.test...