hnp (version 1.2-7)

oil: Diaphorina citri oviposition data

Description

Effects of three agricultural oils on Diaphorina citri oviposition.

Usage

data(oil)

Format

A data frame with 70 observations on the following 2 variables.

y	numeric	number of eggs laid
treat	factor	treatments applied in the experiment

Details

In an experiment to assess the effect of three agricultural oils on the oviposition of Diaphorina citri, seventy Orange Jessamine (Murraya paniculata) plants were sprayed with solutions of the mineral oils Oppa and Iharol, and the vegetable oil Nortox. The experiment used the oils in concentrations of 0.5 and 1.0 percent and a control of plain water set out in a completely randomized design with ten replicates. Following treatment, when the plants were dry, ten pregnant females of D. citri were released on each plant. After five days, the insects were removed and the total number of eggs on each plant was observed, see Amaral et al (2012). This is an example of aggregated data as the number of eggs is the sum over the (unrecorded) numbers of eggs deposited each day and the possibility of day to day variation may contribute additional variability to the recorded counts.

References

Amaral, F. S. A., Poltronieri, A. S., Alves, E. B., Omoto, C. (2012) Efeito de oleos agricolas no comportamento de oviposicao e viabilidade de ovos de Diaphorina citri Kuwayama (Hemiptera: Psyllidae). In: XX Simposio Internacional de Iniciacao Cientifica da Universidade de Sao Paulo, 2012, Pirassununga

Examples

data(oil)

# Poisson fit
model1 <- glm(y ~ treat, family=poisson, data=oil)
anova(model1, test="Chisq")
sum(resid(model1, ty="pearson")^2)

# Quasi-Poisson fit
model2 <- glm(y ~ treat, family=quasipoisson, data=oil)
summary(model2)
anova(model2,test="F")
summary(model2)$dispersion

# Negative binomial fit
require(MASS)
model3 <- glm.nb(y ~ treat, data=oil)
thetahat <- summary(model3)$theta
anova(model3, test="F")

# half-normal plots
par(mfrow=c(1,3),cex=1.4, cex.main=0.9)
hnp(model1,pch=4, main="(a) Poisson",
     xlab="Half-normal scores", ylab="Deviance residuals")
hnp(model2,pch=4, main="(b) Quasi-Poisson",
     xlab="Half-normal scores", ylab='')
hnp(model3,pch=4, main="(c) Negative binomial",
     xlab="Half-normal scores", ylab='')

if (FALSE) {
# using aods3
require(aods3)
model3b <- aodml(y ~ treat, family="nb", phi.scale="inverse",
                 fixpar=list(8, 1.086148), data=oil)
hnp(model3b)
}

## for discussion on the analysis of this data set,
## see Demetrio et al. (2014)