EpiPackage - Childs.Rmd

Packages to use: 	epiR, incidence, epitools, epicalc, Epi
 
```{r}
#install.packages("epicalc", repos = "http://medipe.psu.ac.th/epicalc")

#http://www.columbia.edu/~cjd11/charles_dimaggio/DIRE/resources/R/packages.pdf

#summary tables and statistics - create & visualize
```
```{r}
library(epiR)
library(epitools)
library(incidence)
library(Epi)
library(epiDisplay)
library(epicalc)
library(devtools)
library(roxygen2)
library(plotly)
library(ggplot2)
library(RColorBrewer)
library(AER)

data("Fatalities")

#if comments for when people don't read instructions
test<-Fatalities[complete.cases(Fatalities),]
test1<-na.omit(Fatalities)
```

```{r}
#barplot
ggplot(data=test, aes(x=jail, y=breath)) + geom_bar(stat="identity")

```


```{r}
#boxplot
epi.box<-function(x,y, data, title, xlab, ylab){  #OR epi.box<-function(data, x,y, title, xlab, ylab){
  boxplot(y~x, data=data, notch=TRUE,
  main=title, xlab=xlab, ylab=ylab)
} 
epi.box(Fatalities$jail, Fatalities$income, data=Fatalities, title="main1", xlab="xlab", ylab="ylab")
```

```{r}
#PLOTLY
#sideways hist
plot_ly(Fatalities, y = ~income, color = I("black"), alpha = 0.1, boxpoints = "suspectedoutliers")
#boxplot with scatter
plot_ly(y = ~Fatalities$income, type = "box", boxpoints = "all", jitter = 0.3,pointpos = -1.8) #%>%

#grouped hist
plot_ly(Fatalities,x=~breath, y = ~income, color=~jail, type = "box", boxpoints = "all", jitter = 0.3,
        pointpos = -1.8) %>%
  layout(boxmode = "group")
```



```{r}
  
#using plotly()
 
plot_ly(Fatalities, x = ~income, type = "histogram") 

plot_ly(Fatalities) %>% add_histogram(name = "Income")

#price_hist <- function(method = "FD") {
  #h <- hist(diamonds$price, breaks = method, plot = FALSE)
  #plot_ly(x = h$mids, y = h$counts) %>% add_bars(name = method)
#}

#histogram
plot_ly(Fatalities, x = ~breath, color = ~jail) %>%
  add_histogram()
```

```{r}  
#HISTOGRAM
#histogram with line
ggplot(data=Fatalities, aes(Fatalities$income)) + 
  geom_histogram(aes(y =..density..), 
                 col="blue", 
                 fill="light blue", 
                 alpha=.5) + 
  geom_density(col=2) + 
  labs(title="Histogram for Age", x="Age", y="Count")

#histogram w/o line
library(ggplot2)
ggplot(data=Fatalities, aes(income)) + 
  geom_histogram(col="black",
                 aes(fill=..count..)) +
  scale_fill_gradient("Count", low="light blue", high="navy")+
  labs(title="Histogram for Age", x="Age", y="Count")
```


```{r}

##Vignette for epiplot function

#1. Must have cleaned data with no NAs (Reference link: https://stackoverflow.com/questions/4862178/remove-rows-with-all-or-some-nas-missing-values-in-data-frame )
#2. Arguments: (data, x, y, graph, fill, title, xlab, ylab, legend)
#3. Graphs: bar plot ("bar"), grouped bar plot ("bargroup"), stacked bar plot ("barstack"), box plot ("boxplot"), box plot with correlating dots ("dotboxplot"), histogram ("hist"), density plot ("densityhist"), scatter plot ("scatter"), scatter plot with correlating line ("scatterline"), linear regression ("linreg").
#4. Input the correct arguments:
    #bar plot: (data=data, x=data$x, graph="bar", title="Title of plot", xlab="x-axis label", ylab= "y-axis label")
    #grouped bar plot: (data=data, x=data$x,y=data$y, graph="bargroup",fill=data$fill, title="Title of plot", xlab="x-axis labellab",  ylab="y-axis label")
    #stacked bar plot: (data=data, x=data$x,y=data$y, graph="barstack",fill=data$fill, title="Title of plot", xlab="x-axis labellab",  ylab="y-axis label")
    #box plot: (data=data, x=data$x,y=data$y, graph="boxplot", title="Title of plot", xlab="x-axis label",  ylab="y-axis label")
    #box plot with correlating dots:(data=data,y=data$y, graph="dotboxplot", title="Title of plot", xlab="x-axis labellab",  ylab="y-axis label")
    #histogram: (data=data, x=data$x, graph="hist", title="Title of plot", xlab="x-axis label",  ylab="y-axis label")
    #density plot: (data=data, x=data$x, graph="densityhist", title="Title of plot", xlab="x-axis label",  ylab="y-axis label")
    #scatter plot: (data=data, x=data$x, y=data$y, graph="scatter",fill=data$fill, title="Title of plot", xlab="x-axis label",  ylab="y-axis label", legend="Title of legend fill")
    #scatter plot with correlating line: (data=data, x=data$x, y=data$y, graph="scatterline",fill=data$fill, title="Title of plot", xlab="x-axis label",  ylab="y-axis label", legend="Title of legend fill")
    #linear regression: (data=data, x=data$x, y=data$y, graph="linreg",fill=data$fill, title="Title of plot", xlab="x-axis label",  ylab="y-axis label", legend="Title of legend fill")

```

```{r}
epiplot<- function(data, x, y, graph, fill, title, xlab, ylab, legend){    ###add argument for error bars
  data[complete.cases(data), ]
  if(graph == "bar"){                           #boxplot function
  pic<-  ggplot(data=data, aes(x=x, fill=x)) + 
  geom_bar( ) +
  scale_fill_brewer(palette = "Paired")+
  labs(title="title", x="xlab", y="ylab")
  #table1 = table(data$x)  ## get the cross tab
  #pic<-barplot(table1, beside = TRUE, legend = levels(data$x), col=c("lightblue","darkblue"),main="title", xlab="xlab", ylab = "ylab")  
  return(pic)
  #barplot(table(data$x), col=c("lightblue","darkblue"),main="title", xlab="xlab", ylab = "ylab")
} else if(graph=="bargroup"){                                           #works, but has column for NAs....why???
  pic<-ggplot(data=data, aes(x=x, y=y, fill=fill)) +
  geom_bar(position="dodge", stat="identity") + scale_fill_brewer(palette = "Paired")+theme_bw()+facet_wrap(~"fill")
  return(pic)
} else if(graph=="barstack"){                                           #works, but has column for NAs....why???
  pic<-ggplot(data=data, aes(fill=fill, y=y, x=x)) +
  geom_bar( stat="identity")
  return(pic)
} else if(graph=="boxplot"){                                            #GOOD                 
  pic<-boxplot(y~x, data=data, notch=TRUE,
  main="title", xlab="xlab", ylab="ylab")
  return(pic)
} else if(graph=="dotboxplot"){                                         #GOOD
  pic<-plot_ly(y = ~y, type = "box", boxpoints = "all", jitter = 0.3,pointpos = -1.8) 
  return(pic)
} else if(graph=="hist"){                                               #GOOD
  pic<-ggplot(data=data, aes(x)) +                    
  geom_histogram(col="black", aes(fill=..count..)) +
  scale_fill_gradient("Count", low="light blue", high="navy")+
  labs(title="title", x="xlab", y="ylab")
  return(pic)
} else if(graph=="densityhist"){                                          #GOOD                  
   pic<-ggplot(data=data, aes(x)) + 
  geom_histogram(aes(y =..density..),col="blue", fill="light blue", alpha=.5) + 
  geom_density(col=2) + 
  labs(title="title", x="xlab", y="ylab")
  return(pic)
} else if(graph=="scatter"){                                            #GOOD
  pic<-ggplot(data, aes(x, y, color = fill)) +
  geom_point(shape = 16, size = 5, show.legend = TRUE) +
  theme_minimal() +
  #scale_color_gradient(color = "Blues")+
  labs(title="title", x="xlab", y="ylab", color = "legend")
  return(pic)
} else if(graph=="scatterline"){                                        #GOOD
  pic<-ggplot(data, aes(x, y, color = fill)) +
  geom_point(shape = 16, size = 5, show.legend = TRUE) +
  theme_minimal() +
  scale_color_gradient(low = "light blue", high = "dark blue")+
  labs(title="title", x="xlab", y="ylab", color = "legend")+geom_smooth()
  return(pic)
} else if(graph=="linreg"){                                             #GOOD
  pic<-ggplot(data, aes(x, y, color = fill)) +
  geom_point(shape = 16, size = 5, show.legend = TRUE) +
  theme_minimal() +
  scale_color_gradient(low = "light blue", high = "dark blue")+
  labs(title="title", x="xlab", y="ylab", color = "legend")+ geom_smooth(method = 'lm', se = TRUE)
  return(pic)
} 
}

```

```{r}
epiplot(Fatalities, x=Fatalities$jail, graph = "bar") 

epiplot(Fatalities, y=Fatalities$income, graph = "dotboxplot")

library(ggplot2)
ggplot(Fatalities, aes(Fatalities$income, Fatalities$fatal, color = Fatalities$drinkage)) +
  geom_point(shape = 16, size = 5, show.legend = TRUE) +
  theme_minimal() +
  scale_color_gradient(low = "light blue", high = "dark blue")+
  labs(title="title", x="xlab", y="ylab", color = "Drinkage")

ggplot(Fatalities, aes(Fatalities$income, Fatalities$fatal, color = Fatalities$drinkage)) +
  geom_point(shape = 16, size = 5, show.legend = TRUE) +
  theme_minimal() +
  scale_color_gradient(low = "light blue", high = "dark blue")+
  labs(title="title", x="xlab", y="ylab")+geom_smooth()

###link to complete cases code in stat exchange

class(data1.1$breath)
xlim(Fatalities$breath)
lims(Fatalities$breath)
#Fatalities, aes(income, fatal, color = drinkage
#<- function(data, x, y, graph, fill, title, xlab, ylab, legend){    ###add argument for error bars
```

```{r}

#table1 = table(data$x)  ## get the cross tab
#barplot(table1, beside = TRUE, legend = levels(data$x)) 

#table1 = table(Fatalities$breath)  ## get the cross tab
#barplot(table1, beside = TRUE, legend = levels(Fatalities$breath)) 

#barplot.default(table(data1.1$jail), col = c("lightblue", "darkblue"), main = "title", xlab = "xlab", ylab = "ylab")
  #else if(graph=kaplan){                        #survival
#}
 # print a function with parmfrow
  #ggplot(Fatalities, aes(income, fatal, color = drinkage)) +
  #geom_point(shape = 16, size = 5, show.legend = TRUE) +
  #theme_minimal() +
  #scale_color_gradient(low = "light blue", high = "dark blue")+
  #labs(title="title", x="xlab", y="ylab", color = "Drinks")+geom_smooth(method = 'lm', se = TRUE)
````


```{r}
#test

mtcars<-mtcars
counts <- table(mtcars$gear)
barplot(counts, main="Car Distribution", 
  	xlab="Number of Gears")

fxn1<- function(data, x){    ###add argument for error bars
  table1 <- table(data$x) ## get the cross tab
  barplot(table1)  
}

fxn1(data = mtcars, x=gear)
table1

table1 = table(data1.1$breath)  ## get the cross tab
  pic<-barplot(table1, beside = TRUE, legend = levels(data1.1$breath), col=c("lightblue","darkblue"),main="title", xlab="xlab", ylab = "ylab")  

  #barplot(table(data$x), col=c("lightblue","darkblue"),main="title", xlab="xlab", ylab = "ylab")


is.na(Fatalities$jail)

dat1<-complete.cases(Fatalities)
table(dat1)
```

```{r}


ggplot(data=Fatalities, aes(x=breath, y=income)) +
  geom_bar(stat="identity")+
  theme_bw()+ 
 scale_fill_brewer(palette = "Blues")
  
  barplot(table(data$x), col=c("lightblue","darkblue"),main="title", xlab="xlab", ylab = "ylab")
  
  
plot_ly(Fatalities, x =~income, y = ~drinkage, color = ~income, size = ~income)
  
  ggplot(data, aes(fill=fill, y=y, x=x)) +
    geom_bar(position="dodge", stat="identity") + scale_fill_gradient("Count", low="light blue", high="navy")
  ggplot(data, aes(fill=fill, y=y, x=x)) +
    geom_bar(position="dodge", stat="identity") + scale_fill_brewer(palette = "Blues")+theme_bw()

  
  ggplot(Fatalities, aes(y=income, x=breath, color = jail)) +geom_bar( stat="identity")+ scale_fill_brewer(palette = "Blues")+theme_bw()+ facet_wrap(~breath)

  
  table(Fatalities$jail)
##how to deal with NAs
  
  bars <- c(4, 3, 5, 4.5, 5.1)
labels <-  c('Anthrax', 'DRI', 'DOA', 'BIOHAZARD', 'Withstand')
barplot(bars, names.arg = labels, col = brewer.pal(5, name = "Blues"))

is.na(Fatalities$jail)


plot_ly(Fatalities, x =~income, y = ~fatal, color = ~income, size = ~income)


p<-plot_ly(Fatalities, x =~income, y = ~fatal,  type="scatter", mode = "markers", marker=list( size=20 , opacity=0.5),
        color = ~drinkage, colors=c("light blue","dark blue") %>%
          layout(title = 'Primates Brain and Body Weight',
         xaxis = list(title = 'Body Weight (kg)'),
         yaxis = list(title = 'Brain Weight (g)'))
p

ggplot(Fatalities, aes(income, fatal, color = drinkage)) +
  geom_point(shape = 16, size = 5, show.legend = TRUE) +
  theme_minimal() +
  scale_color_gradient(low = "light blue", high = "dark blue")+
  labs(title="title", x="xlab", y="ylab", color = "Drinks")+geom_smooth()




ggplot(data = dfm, aes(x = TY, y = value, color = variable)) + 
  geom_point(size=5) +
  labs(title = "Temperatures\n", x = "TY [°C]", y = "Txxx", color = "") +
  scale_color_manual(labels = c("T999", "T888"), values = c("blue", "red")) +
  theme_bw() +
  theme(axis.text.x = element_text(size = 14), axis.title.x = element_text(size = 16),
        axis.text.y = element_text(size = 14), axis.title.x = element_text(size = 16),
        plot.title = element_text(size = 20, face = "bold", color = "darkgreen"))


iris, x = ~Petal.Length, y = ~Petal.Width , type="scatter", mode = "markers",
        marker=list( size=20 , opacity=0.5), color = ~Sepal.Length ,
        colors=c("green","blue")



```

```{r}
To include:
  2x2 table
  incidence
  rate ratio
  risk ratio
  odds ratio
  CIs   ###****add CIs???

stargazer pckg
create basic statistics - stargazer to generate
produce stats and put into a table 


shiny app - drag by year - scroll bar by year - define axis as based on the data

broom to tidy into a table

function to produce table with all their measures - beautify tables

r oxygen will generate manual 

write function that formats data the way we want it to - enter column names
naming mechanism for variables and columns - survey makes you identify
function renames dataset with prevalence, incidence - output just anoter dataset

include links in the vignettes - explain and then for future details - insert links

epi.2by2

```
```{r}
#epi.2by2
#run all calculation and then use broom to clean up neatly

dat <- matrix(c(13,2163,5,3349), nrow = 2, byrow = TRUE)
rownames(dat) <- c("DF+", "DF-"); colnames(dat) <- c("FUS+", "FUS-"); dat

epi.2by2(dat = as.table(dat), method = "cross.sectional", 
   conf.level = 0.95, units = 100,  homogeneity = "breslow.day", 
   outcome = "as.columns")

```