FOMC meeting dates, rate changes, scrape with R code

gistfile1.txt

### Setup

```{r, include = FALSE}

library(tidyverse)
library(tidyquant)
library(timetk)
library(scales)
library(janitor)

library(httr)
library(jsonlite)
library(XML)
library(rvest)
library(htmltools)

library(gt)
library(gtExtras)
```

### Introduction

Below is a function to scrape the federal reserve website for fomc meeting dates and rate decisions. 
Note that this takes us back only from 2023 to 2003. The data are here: 

https://www.federalreserve.gov/monetarypolicy/openmarket.htm

For 2002 back to 1990, the Fed stores those at a separate archive html here: 

https://www.federalreserve.gov/monetarypolicy/openmarket_archive.htm

### Build Function

```{r}
fed_funds_change_scraper <- function(chosen_number){
url <- "https://www.federalreserve.gov/monetarypolicy/openmarket.htm"
  
# 1 = 2023
# 14 = 2003

# Read the HTML content of the webpage
webpage <- read_html(url)

html_cell_number <- chosen_number
# Extract the table containing the interest rate change dates
table <- html_nodes(webpage, "table")[[html_cell_number]]

# Extract the rows of the table containing the interest rate change dates
rows <-  html %>% 
  html_nodes("table") %>% 
  .[[1]]

# Extract the dates from the rows and filter for changes since 1970
change_dates <- 
  html_nodes(rows, "td:nth-child(1)") %>%
  html_text() 

increases <- 
  html_nodes(rows, "td:nth-child(2)") %>%
  html_text() 

decreases <- 
  html_nodes(rows, "td:nth-child(3)") %>%
  html_text() %>%
  str_replace("75-100", "85")

year <- html_nodes(webpage, "h4")[[html_cell_number +1]] %>% 
  html_text()

tibble(
  date = str_glue("{change_dates}, {year}") %>% parse_date_time(orders = "%B %d, %Y") %>% ymd(),
  increases = as.numeric(increases)/100,
  decreases = as.numeric(decreases)/-100
)
}

# use map_dfr() to iterate across the html nodes

fed_funds_changes_current <- 
map_dfr(seq(1, 14),
        fed_funds_change_scraper) %>% 
  mutate(change = 
           case_when(
             increases != 0 ~ increases,
             T ~ decreases
           )) %>% 
  select(date, change)

fed_funds_changes_current %>% 
  gt() %>% 
  fmt_percent(change, 
              scale_values = F) %>% 
  gt_theme_538()


```