R/eco.download.R

In soloemoon/ecoMoon: Download, Transform, and visualize economic data

eco.download <- function(source, keys, shape, periodicity, show.recession, time.series){

  # Set default values
  source <-ifelse(missing(source), 'fred', tolower(source))
  periodicity <-ifelse(missing(periodicity),'monthly',tolower(periodicity))
  shape <- ifelse(missing(shape),'wide',tolower(shape))
  show.recession <-ifelse(missing(show.recession),'y',tolower(show.recession))
  time.series <-ifelse(missing(time.series),'n',tolower(time.series))

  # Ensure function inputs have proper classes
  if(inherits(keys, 'character') == FALSE){
    warning('keys entered in improper class')
    return(print('please enter keys as character'))
  }

  if(inherits(shape, 'character') == FALSE){
    warning('Shape entered in improper class')
    return(print('please enter shape as character'))
  }

  if(inherits(show.recession, 'character') == FALSE){
    warning('show.recession entered in improper class')
    return(print('please enter show.recession as character'))
  }

  if(inherits(time.series, 'character') == FALSE){
    warning('time.series entered in improper class')
    return(print('please enter time.series as character'))
  }
  # Check periodicity
  if(inherits(periodicity, 'character') == FALSE){

    warning('Periodicty entered with improper class')
    return(print('Please enter periodicity as a character'))
  }

    if( show.recession=='y'){
      if(periodicity == 'monthly'){
        recession_name <-'USRECM'
        keys <-append(keys, recession_name)
        recession <-fredr('USRECM')
      } else if(periodicity== 'quarterly'){
        recession_name <-'USRECQ'
        keys <-append(keys, recession_name)
        recession <-fredr
      } else if(periodicity== 'daily'){
        recession_name <-'USRECD'
        keys <-append(keys, recession_name)
      }
    }else{
      keys <- keys
    }

  if(source == 'fred'){
    # Download FRED Data
    df <-data.frame(map_dfr(keys,fredr))
    df <-df[order(df$series_id,as.Date(df$date, format="%d/%m/%Y")), ]

  }else if(source == 'bls'){
    bls_data_codes <-list(
      labor_force = 'LNS11000000',
      employment = 'LNS12000000',
      unemployment = 'LNS13000000',
      unemployment_rate = 'LNS14000000',
      nonfarm_employment = 'CES0000000001',
      avg_weekly_hrs = 'CES0500000002',
      avg_weekly_hrs_nonsup = 'CES0500000007',
      avg_hrly_earnings = 'CES0500000003',
      avg_hrly_earnings_nonsup = 'CES0500000008',
      nonfarm_productivity = 'PRS85006092',
      nonfarm_labor_costs = 'PRS85006112',
      nonfarm_hrly_compensation = 'PRS85006152',
      nonfarm_multi_productivity = ' MPU4910012',
      cpi_urban_80s = 'CUUR0000SA0',
      cpi_urban_67 = 'CUUR0000AA0',
      core_cpi_u = 'CUUR0000SA0L1E',
      core_cpi_w = 'CWUR0000SA0L1E',
      ppi_final_sa = 'WPSFD4',
      ppi_final_ua = 'WPUFD4',
      ppi_final_core = 'WPUFD49104',
      ppi_final_core_less_trade = 'WPUFD49116',
      ppi_finshed_goods_82 = 'WPUFD49207',
      imports_all_commodities = 'EIUIR',
      exports_all_commodities = 'EIUIQ',
      employment_cost_index = 'CIU1010000000000A',
      eci_private = 'CIU2010000000000A',
      eci_private_wages = 'CIU2020000000000A'
    )
    # Download dataset
    keys <-unlist(bls_data_codes[keys], use.names = F)
    df <- blsAPI(list('seriesid'=c(keys)) , return_data_frame = T)
    df$date <-as.Date(ydm(paste(df$year, 01, df$periodName)))
    df <-df[c('date', 'seriesID','value')]
    colnames(df) <-c('date', 'series_id','value')

    # Add in join for recession data
  }
# Shape dataset
  if(shape == 'wide'){
    key_col_names <-paste0('value.', keys)
    df <-stats::reshape(df, idvar='date', timevar='series_id', direction='wide')
    rownames(df) <-df$date
    df <-df[ , c('date', key_col_names)]
    names(df)[names(df) == paste0('value.',recession_name)] <- 'recession'
    # Formats data as xts if desired
    if(time.series ==  'y'){
      df$series_id <-NULL
      df <-xts::xts(x = df, order.by = df$date)
      df$date <-NULL
      storage.mode(df) <-'numeric'
    }
  }else if(shape == 'long'){
    df
  }
# Add additional date columns
  df$quarter <-paste(quarters(as.Date(df$date)), format(df$date, '%y'), sep= ' ')
  df$year <-format(df$date, '%y')
  return(df)
}