---
title: "Analyse de l'incience du syndrome grippal"
author: "Olivier Messina"
date: "4/25/2020"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

```{r}
data_url = "https://www.sentiweb.fr/datasets/incidence-PAY-3.csv"
```

```{r}
data = read.csv(data_url, skip=1)
head(data)
```

```{r}
tail(data)
```


```{r}
lignes_na = apply(data,1,function(x) any(is.na(x)))
data[lignes_na,]
```
```{r}
class(data$week)
```
```{r}
class(data$inc)
```
```{r}
data=read.csv(data_url,skip=1, na.strings="-")
```

```{r}
library(parsedate)
```

```{r}
convert_week = function(date) {
  ws = paste(date)
  iso = paste0(substring(ws,1,4),"-W",substring(ws,5,6))
  as.character(parse_iso_8601(iso))
}
```
```{r}
data$date = as.Date(sapply(data$week,convert_week))
```
```{r}
class(data$date)
```

```{r}
data= data[order(data$date),]
head(data)
```
```{r}
all(diff(data$date) ==7)
```
```{r}
with(data,plot(date,inc,type="l"))
```
```{r}
with(tail(data,200),plot(date,inc,type="l"))
```
```{r}
pic_annuel = function(annee){
  debut = paste0(annee-1,"-08-01")
  fin = paste0(annee,"-08-01")
  semaines = data$date > debut & data$date <= fin
  sum(data$inc[semaines],na.rm=TRUE)
}
```
```{r}
annees=1986:2017
```
```{r}
incidence_annuelle = data.frame(annee = annees,
                                incidence = sapply(annees, pic_annuel))
```
```{r}
head(incidence_annuelle)
```
```{r}
plot(incidence_annuelle,type="p")
```
```{r}
incidence_annuelle[order(-incidence_annuelle$incidence),]
```
```{r}
hist(incidence_annuelle$incidence,breaks = 10)
```