glm_binomial.R

## ----knitr_setup, include=FALSE, cache=FALSE----------------------------------------------------------

library("knitr")

### Chunk options ###

## Text results
opts_chunk$set(echo = FALSE, warning = FALSE, message = FALSE, size = 'tiny')

## Code decoration
opts_chunk$set(tidy = FALSE, comment = NA, highlight = TRUE, prompt = FALSE, crop = TRUE)

# ## Cache
# opts_chunk$set(cache = TRUE, cache.path = "knitr_output/cache/")

# ## Plots
# opts_chunk$set(fig.path = "knitr_output/figures/")
opts_chunk$set(fig.align = 'center', out.width = '90%')

### Hooks ###
## Crop plot margins
knit_hooks$set(crop = hook_pdfcrop)

## Reduce font size
## use tinycode = TRUE as chunk option to reduce code font size
# see http://stackoverflow.com/a/39961605
knit_hooks$set(tinycode = function(before, options, envir) {
  if (before) return(paste0("\n \\", options$size, "\n\n"))
  else return("\n\n \\normalsize \n")
  })



## ----prepare_titanic_data, echo=FALSE, eval=FALSE-----------------------------------------------------
## titanic <- read.table("http://www.amstat.org/publications/jse/datasets/titanic.dat.txt")
## names(titanic) <- c("class", "age", "sex", "survived")
## titanic$class <- factor(titanic$class, labels = c("crew", "first", "second", "third"))
## titanic$age <- factor(titanic$age, labels = c("child", "adult"))
## titanic$sex <- factor(titanic$sex, labels = c("female", "male"))
## write.csv(titanic, file = "data/titanic_long.csv", row.names=FALSE, quote=FALSE)


## ----read_titanic, echo=FALSE-------------------------------------------------------------------------
titanic <- read.csv("data/titanic_long.csv")
head(titanic)


## ----titanic_lm, echo=TRUE----------------------------------------------------------------------------
m5 <- lm(survived ~ class, data = titanic)
library("easystats")
check_model(m5)


## ----titanic_lm_resid, echo=FALSE---------------------------------------------------------------------
hist(resid(m5))


## ----out.width="40%"----------------------------------------------------------------------------------
include_graphics("images/modeling_process.png")


## ----echo=TRUE----------------------------------------------------------------------------------------
table(titanic$class, titanic$survived)


## ----titanic_dplyr, echo=c(-1)------------------------------------------------------------------------
library("dplyr")
titanic %>%
  group_by(class, survived) %>%
  summarise(count = n())


## ----titanic_eda, echo=TRUE, out.width = "100%"-------------------------------------------------------
plot(factor(survived) ~ factor(class), data = titanic)


## ----echo=TRUE, out.width="80%"-----------------------------------------------------------------------
library("ggmosaic")
ggplot(titanic) +
  geom_mosaic(aes(x = product(survived, class))) +
  labs(x = "", y = "Survived")


## ----echo=TRUE----------------------------------------------------------------------------------------
tit.glm <- glm(survived ~ class, 
               data = titanic, 
               family = binomial)


## ----titanic_glm, echo=1------------------------------------------------------------------------------
tit.glm <- glm(survived ~ class, data = titanic, family = binomial)
summary(tit.glm)


## ----tit_glm_effects, echo=TRUE-----------------------------------------------------------------------
library("effects")
allEffects(tit.glm)


## ----tit_glm_effects2, echo=TRUE----------------------------------------------------------------------
summary(allEffects(tit.glm))


## ----echo=TRUE----------------------------------------------------------------------------------------
library("easystats")   # 'modelbased' pkg
estimate_means(tit.glm)


## ----echo=TRUE----------------------------------------------------------------------------------------
estimate_contrasts(tit.glm)


## ----echo=TRUE----------------------------------------------------------------------------------------
library("easystats")   # 'performance' pkg
r2(tit.glm)


## ----echo=TRUE----------------------------------------------------------------------------------------
kable(xtable::xtable(tit.glm), digits = 2)


## ----echo=TRUE----------------------------------------------------------------------------------------
library("modelsummary")
modelsummary(tit.glm, output = "markdown")


## ----effects_plot, echo=TRUE--------------------------------------------------------------------------
plot(allEffects(tit.glm))


## ----echo=2-------------------------------------------------------------------------------------------
library("visreg")
visreg(tit.glm, scale = "response", rug = FALSE)


## ----echo=4, out.width="70%", eval=FALSE--------------------------------------------------------------
## library(ggplot2)
## library(sjPlot)
## theme_set(theme_minimal(base_size = 16))
## sjPlot::plot_model(tit.glm, type = "eff")


## ----echo=2-------------------------------------------------------------------------------------------
library("easystats")
plot(parameters(tit.glm), show_intercept = TRUE)


## ----echo=TRUE, out.width="70%"-----------------------------------------------------------------------
no.intercept <- glm(survived ~ class - 1, family = binomial, data = titanic)
plot(parameters(no.intercept))


## ----tit_glm_check, echo=2, tinycode = TRUE-----------------------------------------------------------
layout(matrix(1:4, nrow = 2))
plot(tit.glm)
dev.off()


## ----echo=TRUE, out.width="80%", eval=FALSE-----------------------------------------------------------
## plot(binned_residuals(tit.glm))


## ----binnedplot, eval=FALSE---------------------------------------------------------------------------
## predvals <- predict(tit.glm, type="response")
## arm::binnedplot(predvals, titanic$survived - predvals)


## ----echo=TRUE, eval=FALSE----------------------------------------------------------------------------
## check_model(tit.glm)


## ----out.width="70%", echo = 2, tinycode = TRUE-------------------------------------------------------
library("easystats")
check_predictions(tit.glm)


## ----eval=FALSE---------------------------------------------------------------------------------------
## library(bayesplot)
## sims <- simulate(tit.glm, nsim = 100)
## ppc_bars(titanic$survived, yrep = t(as.matrix(sims)))


## ----echo=TRUE----------------------------------------------------------------------------------------
library("DHARMa")
simulateResiduals(tit.glm, plot = TRUE)


## ----echo=TRUE, out.width="60%"-----------------------------------------------------------------------
library("predtools")
titanic$surv.pred <- predict(tit.glm, type = "response")
calibration_plot(data = titanic, obs = "survived", pred = "surv.pred", 
                 x_lim = c(0,1), y_lim = c(0,1)) 


## ----echo=FALSE---------------------------------------------------------------------------------------
visreg(tit.glm, scale = "response")


## ----tit_sex_eda, out.width="100%"--------------------------------------------------------------------
plot(factor(survived) ~ as.factor(sex), data = titanic)


## ----tit_sex, echo=FALSE------------------------------------------------------------------------------
tit.sex <- glm(survived ~ sex, data = titanic, family = binomial)
summary(tit.sex)


## ----tit_sex_effects, echo=TRUE-----------------------------------------------------------------------
estimate_means(tit.sex)


## ----echo=FALSE, out.width="50%"----------------------------------------------------------------------
visreg(tit.sex, scale = "response", rug = FALSE)


## ----echo=TRUE----------------------------------------------------------------------------------------
simulateResiduals(tit.sex, plot = TRUE)


## ----echo=FALSE, out.width="70%"----------------------------------------------------------------------
library("dagitty")
g1 <- dagitty("dag {
              Class -> Survival
              Sex -> Survival
              Sex -> Class
              }")
plot(g1)


## ----tit_women, echo=TRUE-----------------------------------------------------------------------------
table(titanic$class, titanic$survived, titanic$sex)


## ----echo=FALSE---------------------------------------------------------------------------------------
tit.sex.class.add <- glm(survived ~ class + sex, family = binomial, data = titanic)
summary(tit.sex.class.add)


## ----echo=2-------------------------------------------------------------------------------------------
par(mfcol = c(1, 2))
visreg(tit.sex.class.add, scale = "response", rug = FALSE)
dev.off()


## ----tit_sex_class, echo=FALSE------------------------------------------------------------------------
tit.sex.class.int <- glm(survived ~ class * sex, family = binomial, data = titanic)
summary(tit.sex.class.int)


## ----tit_sex_class_effects, echo=TRUE-----------------------------------------------------------------
estimate_means(tit.sex.class.int)


## ----tit_sex_class_effects2, echo=TRUE----------------------------------------------------------------
visreg(tit.sex.class.int, by = "sex", xvar = "class", scale = "response", rug = FALSE)


## ----echo=TRUE, out.width="80%"-----------------------------------------------------------------------
library("sjPlot")
plot_model(tit.sex.class.int, type = "int")


## ----echo=T-------------------------------------------------------------------------------------------
library("easystats")   # 'performance' pkg
compare_performance(tit.sex.class.add, tit.sex.class.int)


## ----echo=T-------------------------------------------------------------------------------------------
compare_parameters(tit.sex.class.add, tit.sex.class.int)


## ----echo=FALSE, eval = FALSE-------------------------------------------------------------------------
## ## Calibration plot
## titanic$surv.pred <- predict(tit.sex.class.int, type = "response")
## calibration_plot(data = titanic, obs = "survived", pred = "surv.pred",
##                  x_lim = c(0,1), y_lim = c(0,1), nTiles = 10)


## ----read_tit_short, echo = FALSE---------------------------------------------------------------------
tit.prop <- read.csv("data/titanic_prop.csv")
head(tit.prop)


## ----binom_prop, echo=1-------------------------------------------------------------------------------
prop.glm <- glm(cbind(Yes, No) ~ Class, data = tit.prop, family = binomial)
summary(prop.glm)


## ----prop_glm_effects, echo=FALSE---------------------------------------------------------------------
allEffects(prop.glm)


## ----comp, echo=FALSE---------------------------------------------------------------------------------
allEffects(tit.glm)


## ----echo=TRUE----------------------------------------------------------------------------------------
estimate_means(prop.glm)


## ----read_gdp, echo = FALSE---------------------------------------------------------------------------
#gdp <- read.csv("http://vincentarelbundock.github.io/Rdatasets/csv/car/UN.csv")
gdp <- read.csv("data/UN_GDP_infantmortality.csv")
names(gdp) <- c("country", "mortality", "gdp")
summary(gdp)


## ----gdp_eda------------------------------------------------------------------------------------------
plot(mortality ~ gdp, data = gdp, main = "Infant mortality (per 1000 births)")


## ----gdp_glm, echo=1----------------------------------------------------------------------------------
gdp.glm <- glm(cbind(mortality, 1000 - mortality) ~ gdp, 
               data = gdp, family = binomial)
summary(gdp.glm)


## ----gdp_effects, echo=T------------------------------------------------------------------------------
allEffects(gdp.glm)


## ----gdp_effectsplot----------------------------------------------------------------------------------
plot(allEffects(gdp.glm))


## ----gdp_visreg, echo=c(2,3)--------------------------------------------------------------------------
library(visreg)
visreg(gdp.glm, scale = "response")
points(mortality/1000 ~ gdp, data = gdp)


## ----echo=TRUE----------------------------------------------------------------------------------------
simulateResiduals(gdp.glm, plot = TRUE)


## ----echo = TRUE--------------------------------------------------------------------------------------
simres <- simulateResiduals(gdp.glm, refit = TRUE)
testDispersion(simres, plot = FALSE)


## ----logreg_overdisp, echo=1--------------------------------------------------------------------------
gdp.overdisp <- glm(cbind(mortality, 1000 - mortality) ~ gdp, 
               data = gdp, family = quasibinomial)
summary(gdp.overdisp)


## ----logreg_overdisp2, echo=TRUE----------------------------------------------------------------------
parameters(gdp.overdisp)


## ----echo=T-------------------------------------------------------------------------------------------
parameters(gdp.glm)


## ----overdisp_plot1, echo=FALSE, out.width="100%"-----------------------------------------------------
visreg(gdp.glm, scale = "response", main = "Binomial")
points(mortality/1000 ~ gdp, data = gdp, pch = 20)


## ----overdisp_plot2, echo=FALSE, out.width="100%"-----------------------------------------------------
visreg(gdp.overdisp, scale = "response", main = "Quasibinomial")
points(mortality/1000 ~ gdp, data = gdp, pch = 20)


## -----------------------------------------------------------------------------------------------------
visreg(gdp.glm, ylab = "Mortality (logit scale)")


## ----echo=FALSE---------------------------------------------------------------------------------------
library(ggResidpanel)
resid_panel(gdp.glm)


## ----echo=FALSE---------------------------------------------------------------------------------------
gdp.na <- na.omit(gdp)
gdp.na$fit <- fitted(gdp.glm)
plot(gdp.na$fit, gdp.na$mortality, 
     xlab = "Probability of mortality (predicted)",
     ylab = "Mortality (observed)")


## -----------------------------------------------------------------------------------------------------
gdp.na$mort.pred <- predict(gdp.glm, type = "response")
gdp.na$mort.prob <- gdp.na$mortality/1000
calibration_plot(data = gdp.na, obs = "mort.prob", pred = "mort.pred", 
                 x_lim = c(0,0.1), y_lim = c(0,0.1), nTiles = 10)


## ----echo=FALSE, out.width="100%"---------------------------------------------------------------------
visreg(gdp.overdisp, main = "Mortality ~ GDP", ylab = "Mortality (logit scale)")


## ----echo=FALSE, out.width="100%"---------------------------------------------------------------------
gdp.overdisp2 <- glm(cbind(mortality, 1000 - mortality) ~ gdp + I(gdp*gdp), 
               data = gdp, family = quasibinomial)
visreg(gdp.overdisp2, main = "Mortality ~ GDP + GDP^2", ylab = "Mortality (logit scale)")


## ----echo=FALSE, out.width="100%"---------------------------------------------------------------------
visreg(gdp.overdisp, main = "Mortality ~ GDP", scale = "response", ylab = "Mortality")
points(mortality/1000 ~ gdp, data = gdp, pch = 20)


## ----echo=FALSE, out.width="100%"---------------------------------------------------------------------
gdp.overdisp2 <- glm(cbind(mortality, 1000 - mortality) ~ gdp + I(gdp*gdp), 
               data = gdp, family = quasibinomial)
visreg(gdp.overdisp2, main = "Mortality ~ GDP + GDP^2", scale = "response", ylab = "Mortality")
points(mortality/1000 ~ gdp, data = gdp, pch = 20)


## ----echo=FALSE, out.width="100%"---------------------------------------------------------------------
visreg(gdp.overdisp, main = "Mortality ~ GDP", ylab = "Mortality (logit scale)")


## ----echo=FALSE, out.width="100%"---------------------------------------------------------------------
gdp.overdisp2 <- glm(cbind(mortality, 1000 - mortality) ~ log(gdp), 
               data = gdp, family = quasibinomial)
visreg(gdp.overdisp2, main = "Mortality ~ log(GDP)", ylab = "Mortality (logit scale)")


## ----echo=FALSE, out.width="100%"---------------------------------------------------------------------
visreg(gdp.overdisp, main = "Mortality ~ GDP", scale = "response", ylab = "Mortality")
points(mortality/1000 ~ gdp, data = gdp, pch = 20)


## ----echo=FALSE, out.width="100%"---------------------------------------------------------------------
gdp.overdisp2 <- glm(cbind(mortality, 1000 - mortality) ~ log(gdp), 
               data = gdp, family = quasibinomial)
visreg(gdp.overdisp2, main = "Mortality ~ log(GDP)", scale = "response", ylab = "Mortality")
points(mortality/1000 ~ gdp, data = gdp, pch = 20)


## ----eval=FALSE, echo=FALSE---------------------------------------------------------------------------
## ## Trying Poisson
## m <- glm(mortality ~ log(gdp), data = gdp, family = quasipoisson)
## summary(m)
## visreg(m, scale = "response")
## points(mortality ~ gdp, data = gdp)
## 
## gdp.na <- na.omit(gdp)
## gdp.na$fit <- fitted(m)
## plot(gdp.na$fit, gdp.na$mortality,
##      xlab = "Probability of mortality (predicted)",
##      ylab = "Mortality (observed)")


## ----out.width="90%"----------------------------------------------------------------------------------
include_graphics("images/moths.jpg")


## ----echo = 1-----------------------------------------------------------------------------------------
moth <- read.csv("data/moth.csv")
head(moth)


## ----echo = 1-----------------------------------------------------------------------------------------
moth$REMAIN <- moth$PLACED - moth$REMOVED
head(moth)


## -----------------------------------------------------------------------------------------------------
pred.morph <- glm(cbind(REMOVED, REMAIN) ~ MORPH, data = moth, family = binomial)
summary(pred.morph)


## -----------------------------------------------------------------------------------------------------
plot(allEffects(pred.morph))


## -----------------------------------------------------------------------------------------------------
pred.dist <- glm(cbind(REMOVED, REMAIN) ~ DISTANCE, data = moth, family = binomial)
summary(pred.dist)


## -----------------------------------------------------------------------------------------------------
plot(allEffects(pred.dist))


## -----------------------------------------------------------------------------------------------------
pred.int <- glm(cbind(REMOVED, REMAIN) ~ MORPH * DISTANCE, data = moth, family = binomial)
summary(pred.int)


## -----------------------------------------------------------------------------------------------------
plot(allEffects(pred.int))


## ----echo = TRUE--------------------------------------------------------------------------------------
simulateResiduals(pred.int, plot = TRUE)


## ----echo=FALSE---------------------------------------------------------------------------------------
include_graphics("images/tomato.jpg")


## -----------------------------------------------------------------------------------------------------
seed <- readr::read_csv("data/seedset.csv")
head(seed)
seed$plant <- as.factor(seed$plant)


## -----------------------------------------------------------------------------------------------------
plot(seeds ~ ovulecnt, data = seed)


## -----------------------------------------------------------------------------------------------------
plot(seeds ~ pcmass, data = seed)


## ----echo=FALSE---------------------------------------------------------------------------------------
seedm <- glm(cbind(seeds, ovulecnt - seeds) ~ plant, data = seed, family = binomial)
#summary(seedm)
plot(allEffects(seedm))


## ----echo=FALSE---------------------------------------------------------------------------------------
seedm <- glm(cbind(seeds, ovulecnt - seeds) ~ plant + pcmass, data = seed, family = binomial)
#summary(seedm)
plot(allEffects(seedm))


## ----echo = TRUE--------------------------------------------------------------------------------------
soccer <- read.csv("data/soccer.csv")
soccer


## ----echo=FALSE---------------------------------------------------------------------------------------
soccer.mod <- glm(cbind(Scored, Nshots - Scored) ~ GoalkeeperTeam, data = soccer, family = binomial)
visreg(soccer.mod, scale = "response", 
       ylab = "Probability of scoring")