Anthony Garratt and Ivan Petrella, "Commodity Prices and Inflation Risk", Anthony Garratt and Ivan Petrella, Journal of Applied Econometrics, Vol. 37, No. 2, 2022, pp. 392-414. There are eight data files, all of which are zipped in the file gp-data.zip. They are all CSV (ASCII text) files in DOS format. Unix/Linux users should use "unzip -a". (1) Commodity_Spot.csv: contains annual growth rates of real spot commodity prices. (2) Commodity_ CY_1_month.csv: contains commodity price convenience yield data using futures prices 1-month from the current month. (3) Commodity_ CY_12_months.csv: contains commodity price convenience yield data using futures prices 12-months from the current month. (4) PCE_Non-Commodity.csv: contains U.S. PCE price indices. (5) CPI_Non-Commodity.csv: contains U.S. CPI price indices. (6) EX_Rate_Non-Commodity.csv: contains trade weighted exchange rate index. (7) Fed_Funds_Non-Commodity.csv: contains Fed Funds rate. (8) Unemployment_Non-Commodity.csv: contains the unemployment rate. DATA DESCRIPTION (A) COMMODITY DATA (Commodity_Spot.csv, Commodity_ CY_1_month.csv and Commodity_ CY_12_month.csv) The commodity spot and convenience yield data are constructed from futures contracts published from the relevant exchanges, which are freely accessible from the web provider Quandl: https://www.quandl.com/collections/futures. The sample period is 1984m1 to 2016m4 (388 observations). We consider 23 commodities among the energy, food, grains, industrials, metals and livestock commodity classes. The 23 commodity groups are: Crude Oil, Gas Oil, Heating Oil, Unleaded Gas, Natural Gas, Cocoa, Orange Juice, Coffee, Sugar, Soybean Oil, Corn, Oats, Soybeans, Soybean Meal, Wheat, Cotton, Copper, Lumber, Gold, Silver, Feeder Cattle, Live, Cattle and Lean Hogs. (B) PCE AND CPI PRICE INDICES USED TO DEFINE INFLATION (PCE_Non-Commodity.csv, CPI_Non-Commodity.csv) The results reported in the main body of the paper use the U.S. monthly Personal Consumption Expenditures (PCE) price index to define inflation. Two PCE price indices are used: (i) all items index used to define "headline" inflation and (ii) all items index excluding food and energy used to define "core" inflation. Both are Chain-type price indices, with base year 2012=100, and are seasonally adjusted. Source: Federal Reserve Economic Database (FRED), Federal Reserve Bank of St Louis, https://fred.stlouisfed.org/. Sample period: 1990M1- 2016m12, 324 observations. The data are in PCE_Non-Commodity.csv. In addition, the appendix reports results using the Consumer Price Index (CPI) to define inflation. As with PCE we use two CPI indices, for all items and all items excluding food and energy. The details are as described above for PCE, except the base year(s) are 1982-1984=100. The data are in PCE_CPI_Non-Commodity.csv. (C) NON-COMMODITY BASED CONDITIONING VARIABLES (EX_rate_Non-Commodity.csv,Fed_Funds_Non-Commodity.csv, Unemployment_Non-Commodity.csv) Three non-commodity-based conditioning variables are used in the regressions: (i) A trade-weighted US exchange rate, index 1973=100. Source: FRED. Sample period: 1990M1- 2016m12, 324 observations. The data are in EX_Rate_Non-Commodity.csv. (ii) The effective federal funds interest rate, annual percent, Source: FRED. Sample period: 1990M1- 2016m12, 324 observations. The data are in Fed_Funds_Non-Commodity.csv. (iii) The unemployment Rate (SA), percent, not seasonally adjusted. Source: FRED. Sample period: 1990M1- 2016m12, 324 observations. The data are in Unemployment_Non-Commodity.csv.