00-data-preparation.Rmd

---
title: "Data Wrangling"
output: 
  html_document:
    toc: true
    toc_float: true
    fig_width: 12
    fig_height: 10
    code_folding: hide
editor_options: 
  chunk_output_type: inline
---

```{r message=FALSE, warning=FALSE}
library(readr, warn.conflicts = FALSE)
library(lubridate, warn.conflicts = FALSE)
library(skimr)
library(tibble)
```

# Data loading 

Create csv. file path. 

```{r infile-path, echo=TRUE, message=FALSE, warning=FALSE}
# use here package
raw_data_path <- file.path("../data/raw/")
infile_name   <- "malaria-monthly-cases-district-loreto"
infile        <- paste(infile_name, ".csv", sep = "")
infile_path   <- file.path(raw_data_path, infile)
```

```{r outfile_path}
processed_data_path <- file.path("../data/processed/")
outfile_name        <- "malaria-monthly-cases-district-loreto"
outfile             <- paste("pro_", outfile_name, ".csv", sep = "")
outfile_path        <- file.path(processed_data_path, outfile)
```

Define data types for columns.

```{r column-names, message=FALSE, warning=FALSE}
# Column names
col_names <- c(
  "district",
  "year",
  "month",
  "falciparum",
  "vivax",
  "aet",
  "prcp",
  "q",
  "soilm",
  "tmax",
  "tmin",
  "water_deficit",
  "loss",
  "loss_km2",
  "cum_loss_km2",
  "diag",
  "enviro",
  "nets",
  "workers",
  "pamafro",
  "pop2015",
  "province",
  "region",
  "id_district"
)
```

```{r column-types}
# Column types
col_types <- 
  readr::cols(
    district      = col_character(),
    year          = col_integer(),
    month         = col_integer(),
    falciparum    = col_integer(),
    vivax         = col_integer(),
    aet           = col_double(),
    prcp          = col_double(),
    q             = col_double(),
    soilm         = col_double(),
    tmax          = col_double(),
    tmin          = col_double(),
    water_deficit = col_double(),
    loss          = col_double(),
    loss_km2      = col_double(),
    cum_loss_km2  = col_double(),
    diag          = col_integer(),
    enviro        = col_integer(),
    nets          = col_integer(),
    workers       = col_integer(),
    pamafro       = col_integer(),
    pop2015       = col_integer(),
    province      = col_character(),
    region        = col_character(),
    id_district   = col_character()
  )
```

Reading csv. file.

```{r read-file, message=FALSE, warning=FALSE}
dataset <- 
  readr::read_csv(
    file      = infile_path,
    col_names = col_names,
    col_types = col_types,
    skip      = 1,
    locale    = locale(encoding = "UTF-8")
  )
```

## Data validation 

Inspect data.

```{r inspect, include=FALSE}
# View(dataset)
```

```{r}
head(dataset)
```

Check data structure.

```{r data-structure, include=FALSE, message=FALSE, warning=FALSE}
str(dataset)
```

Summarize

```{r summary}
skimr::skim(dataset)
```

```{r message=FALSE, warning=FALSE}
# TODO Potential function
cat_cols <- c()

for (col in names(dataset)) {
  if (is.character(dataset[[col]])) {
    cat_cols <- c(cat_cols, col)
  }
}

for (col in cat_cols) {
  cat(
    "-",
    col, 
    length(unique(dataset[[col]])), 
    "\n",
    unique(dataset[[col]]),
    "\n"
  )
}
```

# Data preparation

```{r replace-province}
processed_dataset <- tibble::tibble(dataset)

processed_dataset$province <- replace(
  dataset$province,
  dataset$province == "DATEN DEL MARAÑON",
  "DATEM DEL MARAÑON"
)
```

```{r check-replace}
unique(processed_dataset$province)
```

Create a reporting date features in datetime format using `year` and `month` columns.

```{r make-datetime, message=FALSE, warning=FALSE}
processed_dataset$dttm <- lubridate::make_datetime(
  year  = dataset$year,
  month = dataset$month,
  day   = 1L
)
```

```{r encode-dttm}
processed_dataset$time <- as.numeric(processed_dataset$dttm)
```

Column subset for analysis

```{r}
col_analysis <- c(
  "district",
  "year",
  "month",
  "falciparum",
  "vivax",
  "aet",
  "prcp",
  "q",
  "soilm",
  "tmax",
  "tmin",
  "pop2015",
  "province",
  "region",
  "dttm",
  "time"
)

processed_dataset <- processed_dataset[, col_analysis]
```

Inspect processed data set

```{r}
head(processed_dataset)
```

# Export processed data

```{r}
readr::write_csv(processed_dataset, outfile_path)
```