---
title: "Analyse de l'incidence du syndrome grippal"
author: "Clément MOLINA"
date: "2023-08-31"
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?v=4bf55"
```

```{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}
lignes_na = apply(data, 1, function (x) any(is.na(x)))
data [lignes_na,]
```
```{r}
library(parsedate)
```
```{r}
convert_date = function(w) {
  ws = paste(w)
  iso = paste0(substring(ws,1,4),"-W",substring(ws,5,6))
  as.character(parse_iso_8601(iso))
}
```

```{r}
data$date = as.Date(convert_date(data$week))
```

```{r}
class(data$date)
```
```{r}
data = data[order(data$date),]
```


```{r}
all(diff(data$date) == 7)
```
```{r}
plot(data$date, data$inc, type="l", xlab="Date", ylab="Incidence hebdomadaire")
```

```{r}
with(tail(data, 200), plot(date, inc, type="l", xlab="Date", ylab="Incidence hebdomadaire"))
```

```{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}
inc_annuelle = data.frame(annee = 1986:2023,
                          incidence = sapply(1986:2023, pic_annuel))
head(inc_annuelle)

```

```{r}
plot(inc_annuelle, type="p", xlab="Année", ylab="Incidence annuelle")
```

```{r}
tail(inc_annuelle[order(-inc_annuelle$incidence),])
```

```{r}
hist(inc_annuelle$incidence, breaks=10, xlab="Incidence annuelle", ylab="Nb d'observations", main="")
```