---
title: "BOJ"
author: "Stefan Angrick"
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{BOJ}
  %\VignetteEngine{knitr::rmarkdown}
  \usepackage[utf8]{inputenc}
---

```{r setup, echo = FALSE}
knitr::opts_chunk$set(
  message = FALSE,
  warning = FALSE,
  collapse = TRUE,
  fig.width = 6, 
  fig.height = 4
)
```

The `BOJ` package provides an `R` interface to [Bank of Japan](https://www.boj.or.jp/) statistics, specifically the [flat files](https://www.stat-search.boj.or.jp/info/dload_en.html) available on the [BOJ Time-Series Data](https://www.stat-search.boj.or.jp/) portal.

## Import data

To import data, first load the package:

```{r loading}
library("BOJ")
```

Next, run the `get_boj_datasets()` function to obtain a list of available data sets:

```{r datasets, message=FALSE, warning=FALSE}
datasets <- get_boj_datasets()
datasets
```

The function returns a [tibble](https://tibble.tidyverse.org/) data frame listing the available data sets. The column `url` can be used as input for the function `get_boj()` which downloads, parses and imports the corresponding data.

To import monthly-frequency data on Japan's [Services Producer Price Index](https://www.boj.or.jp/en/statistics/pi/sppi_2015/index.htm/), run:

```{r sppi, message=FALSE, warning=FALSE}
sppi <- get_boj(datasets$url[(datasets$name == "sppi_m_en")])
sppi
```

To plot the data using [ggplot2](https://ggplot2.tidyverse.org), run the following:

```{r plot, message=FALSE, warning=FALSE}
library("dplyr")
library("ggplot2")
library("zoo")

sppi_plot <- subset(sppi, code %in% c("PRCS15_5200000000", "PRCS15_5200010001",
                                      "PRCS15_5200010002", "PRCS15_5200010003",
                                      "PRCS15_5200010004", "PRCS15_5200010005",
                                      "PRCS15_5200010006", "PRCS15_5200010007"))
sppi_plot <- mutate(sppi_plot, date = as.Date(as.yearmon(date, format = "%Y%m")))
sppi_plot <- mutate(sppi_plot, struc = gsub("^Major group/ ", "", struc))
sppi_plot <- subset(sppi_plot, !is.na(obs_value))

ggplot(sppi_plot, aes(x = date, y = obs_value)) +
  geom_line(aes(colour = struc)) +
  labs(x = "Date", y = "Services Producer Price Index (2015 base)") +
  theme(legend.title = element_blank())
```

Note that BOJ data sets come with a number of different time formats. The [zoo](https://cran.r-project.org/package=zoo) package (e.g. `as.yearmon()`) should be able to parse most formats.

## Note

This package is in no way officially related to or endorsed by the [Bank of Japan](https://www.boj.or.jp/). It was inspired by the [BIS R package](https://github.com/expersso/BIS). Please don't abuse the BOJ's servers with unnecessary calls.