inflation_data {hdflex}

Dataset to estimate quarterly U.S. inflation

Description

A novel, high-dimensional dataset built by Koop and Korobilis (2023) that merges predictive signals from several mainstream aggregate macroeconomic and financial datasets. The dataset includes the FRED-QD dataset of McCracken and Ng (2020), augment with portfolio data used in Jurado et al. (2015), stock market predictors from Welch and Goyal (2008), survey data from University of Michigan consumer surveys, commodity prices from the World Bank’s Pink Sheet database, and key macroeconomic indicators from the Federal Reserve Economic Data for four economies (Canada, Germany, Japan, United Kingdom). The data is already pre-processed to perform one-step-ahead forecasts and augmented with (external) point forecasts from Koop & Korobilis (2023). The dataset spans the period 1960-Q3 to 2021-Q4.

Usage

inflation_data

Format

A matrix with 245 quarterly observations (rows) and 516 variables (columns).

Column 1:4

Transformed target variables: GDP deflator (GDPCTPI), PCE deflator (PCECTPI), Total CPI (CPIAUCSL), Core CPI (CPILFESL)

Column 5:8

First lag of the target variables

Column 9:12

Second lag of the target variables

Column 13:16

All four (lagged) price series transformed with second differences of logarithms

Column 17:452

All remaining (lagged and transformed) signals from the FRED-QD dataset of McCracken and Ng (2020), portfolio data used in Jurado et al. (2015), stock market predictors from Welch and Goyal (2008), survey data from University of Michigan consumer surveys, commodity prices from the World Bank’s Pink Sheet database, and key macroeconomic indicators from the Federal Reserve Economic Data for Canada, Germany, Japan & United Kingdom.

Column 453:468

External point forecasts for quarterly GDP deflator (GDPCTPI) generated by the MatLab Code from Koop and Korobilis (2023). The forecasts were generated out-of-sample from 1976-Q1 to 2021-Q4.

Column 469:484

External point forecasts for quarterly PCE deflator (PCECTPI) generated by the MatLab Code from Koop and Korobilis (2023). The forecasts were generated out-of-sample from 1976-Q1 to 2021-Q4.

Column 485:500

External point forecasts for quarterly Total CPI (CPIAUCSL) generated by the MatLab Code from Koop and Korobilis (2023). The forecasts were generated out-of-sample from 1976-Q1 to 2021-Q4.

Column 501:516

External point forecasts for quarterly Core CPI (CPILFESL) generated by the MatLab Code from Koop and Korobilis (2023). The forecasts were generated out-of-sample from 1976-Q1 to 2021-Q4.

Source

<https://doi.org/10.1111/iere.12623>

References

Jurado, K., Ludvigson, S. C., and Ng, S. (2015) "Measuring uncertainty." American Economic Review, 105 (3): 1177–1216.

Koop, G. and Korobilis, D. (2023) "Bayesian dynamic variable selection in high dimensions." International Economic Review.

McCracken, M., and S. Ng (2020) “FRED-QD: A Quarterly Database for Macroeconomic Research” National Bureau of Economic Research, Working Paper 26872.

Welch, I. and Goyal, A. (2008) "A comprehensive look at the empirical performance of equity premium prediction." The Review of Financial Studies, 21 (4): 1455–1508.