Readings
A curated collection of readings on trimmed mean inflation and related research.
Median Price Changes: An Alternative Approach to Measuring Current Monetary Inflation
Bryan, Michael F., and Christopher Pike, 1991 — Federal Reserve Bank of Cleveland, Economic Commentary, December 1, 1991
This Commentary considers how fluctuations in the distribution of prices changes distort the interpretation of the inflation trend. The median of the price change distribution minimizes, in a non subjective way, the influence of these transitory price movements.
Bryan, Michael F., and Stephen G. Cecchetti, 1994 — NBER Chapters, in: Monetary Policy, pages 195–219, National Bureau of Economic Research, Inc.
This paper shows that the inflation trend can be measured by a limited-influence estimator such as the median of the cross-sectional distribution of price changes. It demonstrates that when the distribution of sector-specific shocks is skewed, the mean of price changes is a poor measure of persistent inflation. Trimmed means, which average only the central part of the distribution after removing outliers, provide a more accurate measure of the central tendency of inflation.
Efficient Inflation Estimation
Bryan, Michael F., Stephen G. Cecchetti, and Rodney L. Wiggins II, 1997 — NBER Working Paper 6183
This paper investigates the use of trimmed means as high-frequency estimators of inflation. The known characteristics of price change distributions, specifically the observation that they generally exhibit high levels of kurtosis, imply that simple averages of price data are unlikely to produce efficient estimates of inflation. Trimmed means produce superior estimates of ‘core inflation,’ which we define as a long-run centered moving average of CPI and PPI inflation.
Measuring Short-Run Inflation for Central Bankers
Cecchetti, Stephen G., May/June 1997 — Federal Reserve Bank of St. Louis Review 79, no. 3: 143–55
This paper provides a broad overview of the issues of noise and bias in inflation measurement for Central bankers. It compares the performance of trimmed-mean estimators, traditional core measures, and a dynamic factor measure of inflation.
A Robust Measure of Core Inflation in New Zealand, 1949–96
Roger, Scott, 1997 — Reserve Bank of New Zealand Discussion Paper Series G97/7, Reserve Bank of New Zealand, Wellington, NZ
This paper constructs asymmetric trimmed mean and median inflation estimators for quarterly CPI data in New Zealand. It documents the persistent skewness in the New Zealand CPI data and reports the efficiency gains from trimming.
The Monthly Measurement of Core Inflation in Japan
Bryan, Michael F., and Stephen G. Cecchetti, 1999 — Monetary and Economic Studies 17, no. 1 (May): 77–101
This paper documents the distributional properties of monthly CPI component price changes in Japan and constructs trimmed mean estimators of core inflation for Japan.
Inflation and the Distribution of Prices
Bryan, Michael F., and Stephen G. Cecchetti, 1999 — The Review of Economics and Statistics 81, no. 2 (May): 188–96
This paper reconsiders the empirical evidence connecting inflation to its higher-order moments. In particular, it examines the statistical properties of the observed positive correlation between the sample mean and the sample cross-sectional skewness of price changes that has attracted substantial attention. It shows that the sample mean-skewness correlation suffers from a small-sample bias accounting for the entirety of the observed correlation. In other words, one of the stylized facts in the literature on aggregate price behavior need not be a fact at all.
A Note on the Efficient Estimation of Inflation in Brazil
Bryan, Michael F., and Stephen G. Cecchetti, 2001 — Banco Central do Brasil Working Paper No. 11
This paper documents the distributional characteristics of price change distributions for 11 nations, including Brazil. It documents the extreme excess kurtosis and persistent skewness in Brazilian data and constructs an asymmetric trimmed mean and median.
Dolmas, Jim, 2005 — Federal Reserve Bank of Dallas Working Paper No. 0506, Federal Reserve Bank of Dallas
This paper constructs the asymmetric trimmed mean estimator using PCE data and is the foundation for the trimmed mean PCE inflation rate constructed by the Federal Reserve Bank of Dallas.
A Phillips Curve with Anchored Expectations and Short-Term Unemployment
Ball, Laurence M., and Sandeep Mazumder. 2019 — Journal of Money, Credit and Banking 51 (1): 111–37.
Ball and Mazumder specify a simple Phillips curve based on the assumptions that inflation expectations are fully anchored at the Federal Reserve’s target, and that labor-market slack is captured by the level of short-term unemployment. Their equation explains inflation behavior since 2000, including the failure of high total unemployment since 2008 to reduce inflation greatly. The fit of their model is especially good when using the weighted median CPI.