Efficient Inflation Estimation

Abstract

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, We find that trimming 9% from each tail of the CPI price-change distribution, or 45% from the tails of the PPI price-change distribution, yields an efficient estimator of core inflation for these two series, although lesser trims also produce substantial efficiency gains. Historically, the optimal trimmed estimators are found to be nearly 23% more efficient (in terms of root-mean-square error) than the standard mean CPI, and 45% more efficient than the mean PPI. Moreover, the efficient estimators are robust to sample period and to the definition of the presumed underlying long-run trend in inflation.

Tracking Trend Inflation: Nonseasonally Adjusted Variants of the Median and Trimmed-Mean CPI

“…1 See Bryan and Cecchetti (1993, 1994, and 1995, Cecchetti (1997), and Bryan, Cecchetti and Wiggins (1997). 2 Detmeister (2011) is the paper most closely related to this study, and it reaches many similar conclusions regarding the PCE price index.…”

“…Most of the research studies have focused upon forecasting ability. Bryan and Cecchetti theselves conducted numerous follow-up studies, sometimes with coauthors (see, e.g., Cecchetti 1994, andWiggins 1997), and these researchers, along with others, developed similar measures for other countries as well (see, e.g., Cecchetti 1999 andAucremanne 2000). Results have been overwhelmingly positive.…”

Forecasting Distribution of Inflation Rates: Functional Autoregressive Approach

“…Hansen, 1994;Bryan et al, 1997;Harvey and Siddique, 2000;Roger, 2000;Jondeau and Rockinger, 2003). In the case of inflation, Bryan et al (1997) document fat-tailed properties and Roger (2000) provides evidence of rightward skewness. It follows that analysis based on aggregate data may be severely biased given the presence of informational heterogeneity across the sectors as documented by Pesaran and Smith (1995) and Hsiao et al (2005).…”

“…A number of studies have examined the role of time-varying higher-order moments in the analysis of financial and macroeconomic data (e.g. Hansen, 1994;Bryan et al, 1997;Harvey and Siddique, 2000;Roger, 2000;Jondeau and Rockinger, 2003). In the case of inflation, Bryan et al (1997) document fat-tailed properties and Roger (2000) provides evidence of rightward skewness.…”

Measuring Core Inflation by Multivariate Structural Time Series Models

“…Other measures exclude mortgage interest costs, and some also exclude the changes in indirect taxes. Bryan and Cecchetti (1994) (see also Bryan, Cecchetti and Wiggins II, 1997) argue that the systematic exclusion of specific items, such as food and energy, is arbitrary, and, after remarking that the distribution of relative price changes exhibits skewness and kurtosis, propose to use the median or the trimmed mean of the cross-sectional distribution.…”