realtalk makes it easy to use common US price indexes in R.
Loading the data
realtalk is mostly a data package and includes several datasets of common US price indexes, like the CPI-U-RS:
library(realtalk)
cpi_u_rs_annual
#> # A tibble: 48 × 2
#> year cpi_u_rs
#> <dbl> <dbl>
#> 1 1978 104.
#> 2 1979 114.
#> 3 1980 127.
#> 4 1981 139.
#> 5 1982 148.
#> 6 1983 154.
#> 7 1984 160.
#> 8 1985 166.
#> 9 1986 168.
#> 10 1987 174.
#> # ℹ 38 more rowsYou can see what datasets are available by loading available_price_indexes:
available_price_indexes
#> # A tibble: 20 × 6
#> index_name frequency seasonal min_date max_date package_data_name
#> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 C-CPI-U annual <NA> 2000 2025 c_cpi_u_annual
#> 2 C-CPI-U monthly NSA Dec 1999 Oct 2026 c_cpi_u_monthly_nsa
#> 3 C-CPI-U quarterly NSA 2000q1 2026q3 c_cpi_u_quarterly_nsa
#> 4 C-CPI-U, extended annual <NA> 1937 2025 c_cpi_u_extended_annu…
#> 5 C-CPI-U, extended monthly NSA Jan 1937 Oct 2026 c_cpi_u_extended_mont…
#> 6 C-CPI-U, extended monthly SA Jan 1947 Oct 2026 c_cpi_u_extended_mont…
#> 7 C-CPI-U, extended quarterly NSA 1937q1 2026q3 c_cpi_u_extended_quar…
#> 8 C-CPI-U, extended quarterly SA 1947q1 2026q3 c_cpi_u_extended_quar…
#> 9 CPI-U annual <NA> 1937 2025 cpi_u_annual
#> 10 CPI-U monthly NSA Jan 1937 Oct 2026 cpi_u_monthly_nsa
#> 11 CPI-U monthly SA Jan 1947 Oct 2026 cpi_u_monthly_sa
#> 12 CPI-U quarterly NSA 1937q1 2026q3 cpi_u_quarterly_nsa
#> 13 CPI-U quarterly SA 1947q1 2026q3 cpi_u_quarterly_sa
#> 14 CPI-U-RS annual <NA> 1978 2025 cpi_u_rs_annual
#> 15 CPI-U-RS monthly NSA Dec 1977 Dec 2025 cpi_u_rs_monthly_nsa
#> 16 CPI-U-X1 annual <NA> 1967 1982 cpi_u_x1_annual
#> 17 CPI-U-X1 monthly NSA Jan 1967 Dec 1982 cpi_u_x1_monthly_nsa
#> 18 PCE annual <NA> 1929 2025 pce_annual
#> 19 PCE monthly SA Jan 1959 Sep 2026 pce_monthly_sa
#> 20 PCE quarterly SA 1947q1 2026q3 pce_quarterly_saThe package_data_name column contains the internal name of each dataset:
pce_monthly_sa
#> # A tibble: 813 × 3
#> year month pce
#> <int> <int> <dbl>
#> 1 1959 1 15.2
#> 2 1959 2 15.2
#> 3 1959 3 15.2
#> 4 1959 4 15.2
#> 5 1959 5 15.2
#> 6 1959 6 15.3
#> 7 1959 7 15.3
#> 8 1959 8 15.3
#> 9 1959 9 15.4
#> 10 1959 10 15.4
#> # ℹ 803 more rowsYou can also use get_price_index():
get_price_index("PCE", "monthly", "SA")
#> # A tibble: 813 × 3
#> year month pce
#> <int> <int> <dbl>
#> 1 1959 1 15.2
#> 2 1959 2 15.2
#> 3 1959 3 15.2
#> 4 1959 4 15.2
#> 5 1959 5 15.2
#> 6 1959 6 15.3
#> 7 1959 7 15.3
#> 8 1959 8 15.3
#> 9 1959 9 15.4
#> 10 1959 10 15.4
#> # ℹ 803 more rowsInstallation
Install realtalk from r-universe:
install.packages(
"realtalk",
repos = c("https://economic.r-universe.dev", getOption("repos"))
)If you add the EPI r-universe repository to your .Rprofile with
# Example .Rprofile
options(repos = c(
EPI = "https://economic.r-universe.dev",
getOption("repos")
))then you can simply
install.packages("realtalk")Updates
realtalk is usually updated one-to-two times a month, after the Bureau of Labor Statistics publishes new CPI data and the Bureau of Economic Analysis publishes new PCE data.
When you load realtalk, it tells you if its data is two or more months old and a newer version is available. Turn this off with options(realtalk.update_check = FALSE).