R/rhodium.R

In abresler/govtrackR: Federal government tracking tools

.dict_rhodium_names <- function() {
  tibble(
    nameRhodium = c(
      "month",
      "year",
      "millions",
      "direction",
      "millionsPerQuarter",
      "stake",
      "type",
      "investorOwnership",
      "mode",
      "industry",
      "targetKey",
      "sourceKey",
      "sourceCountry"
    ),
    nameActual = c(
      "monthTransaction",
      "yearTransaction",
      "amountInvestment",
      "typeDirection",
      "amountInvestmentCumulativeQuarter",
      "typeStake",
      "typeStrategy",
      "typeInvestor",
      "typeInvestment",
      "sectorInvestment",
      "stateInvestment",
      "stateInvestor",
      "countryInvestor"
    )

  )
}

# chinese_fdi -------------------------------------------------------------


## https://www.us-china-investment.org/us-china-foreign-direct-investments/getData

#' Rhodium group China-US investment tracker
#'
#' Acquires cross border investment data from Rhodium Group US-China investment tracker
#'
#'
#' @param snake_names if `TRUE` return snake case names
#' @param unformat if `TRUE` unformat currency columns
#' @param return_message if `TRUE` unformat columns
#'
#' @return \code{tibble()}
#' @family china
#' \url{https://www.us-china-investment.org}
#' @export
#'
#' @examples
#' \dontrun{
#' rhodium_chinese_fdi()
#' }
rhodium_chinese_fdi <-
  function(snake_names = F,
           unformat = F,
           return_message = T) {
    url = "https://www.us-china-investment.org/us-china-foreign-direct-investments/getData"
    json_data <- fromJSON(url, simplifyDataFrame = T, flatten = T)
    industries <- json_data$filters$industry
    data <- json_data$docs %>% as_tibble()
    names(data) <- names(data) %>% str_remove_all("rowParams.")
    data <- data %>%
      mutate(idRow = 1:n()) %>%
      select(idRow, everything())
    df_q <- data %>%
      select(idRow, quarters)

    df_q <- seq_along(1:nrow(df_q)) %>%
      future_map_dfr(function(x) {
        df_q %>% dplyr::slice(x) %>%
          unnest() %>%
          mutate_all(as.numeric)
      })

    df_q <- df_q %>% group_by(idRow) %>% dplyr::slice(1) %>% ungroup()
    data <-
      data %>% select(-quarters) %>% left_join(df_q, by = "idRow") %>% select(one_of(names(df_q)), everything()) %>%
      arrange(desc(year)) %>%
      select(-idRow)
    dict_names <- .dict_rhodium_names()
    actual_names <- names(data) %>%
      map_chr(function(name) {
        dict_names %>% filter(nameRhodium == name) %>% pull(nameActual)
      })

    data <- data %>%
      setNames(actual_names) %>%
      mutate(isGreenfield = typeInvestment == "Greenfield") %>%
      mutate_at(c("amountInvestmentCumulativeQuarter", "amountInvestment"),
                list(function(x) {
                  x * 1000000 %>% currency(digits = 0)
                }))

    data <-
      data %>%
      mutate(
        dateTransaction = glue("{yearTransaction}-{monthTransaction}-01") %>% ymd() %m+% months(1) - 1
      ) %>%
      select(
        dateTransaction,
        sectorInvestment,
        typeDirection,
        typeStake,
        typeInvestment,
        typeInvestor,
        countryInvestor,
        stateInvestment,
        stateInvestment,
        amountInvestment,
        everything()
      )

    if (return_message) {
      t_count <- nrow(data) %>% formattable::comma(digits = 0)
      total_inv <- data$amountInvestment %>% sum()
      us_count <-
        data %>% filter(countryInvestor == "usa") %>% nrow() %>% comma(digits = 0)
      us_amt <-
        data %>% filter(countryInvestor == "usa") %>% pull(amountInvestment) %>% sum()

      china_count <-
        data %>% filter(countryInvestor == "china") %>% nrow() %>% comma(digits = 0)
      china_amt <-
        data %>% filter(countryInvestor == "china") %>% pull(amountInvestment) %>% sum()
      start <- data$dateTransaction %>% min()
      end <- data$dateTransaction %>% max()
      glue(
        "{t_count} Chinese/US investments totaling {total_inv} from {start} to {end}\n\n{us_count} US to China investments totaling {us_amt}\n\n{china_count} China to US investments totaling {china_amt}"
      ) %>% message()
    }

    data <- data %>%
      munge_data(snake_names = snake_names, unformat = unformat)

    data


  }