Using the Retail Distribution of Sellers to Impute Expenditure Shares

Principal Investigator(s): View help for Principal Investigator(s) Mingzhi Xu, Peking University; Robert C Feenstra, University of California Davis; Alexis Antoniades, Georgetown University Qatar

Version: View help for Version V1

Name	File Type	Size	Last Modified
Code diagram.pdf	application/pdf	278.1 KB	02/21/2022 04:22:AM
Figure 1 Measures of Retail Distribution and Market Share using Chinese Data.do	text/plain	3.4 KB	02/24/2022 03:01:AM
Figure 2 Facts on Retailers Heterogeneity, United Arab Emirates.do	text/plain	6.8 KB	02/23/2022 08:36:PM
Figure 3 Frequency of price changes and market share.do	text/plain	2.4 KB	02/22/2022 09:16:AM
Figure 4 Frequency of price changes by product module US.do	text/plain	838 bytes	02/23/2022 03:29:AM
README.pdf	application/pdf	641.1 KB	07/24/2022 07:06:AM
Table 1 Descriptive Statistics.do	text/plain	1020 bytes	02/21/2022 07:06:PM
Table 2 regressions ND WD GCC.do	text/plain	1022 bytes	02/21/2022 10:42:PM
Table 2 regressions ND WD US.do	text/plain	1.7 KB	02/23/2022 08:04:PM
Table 3 Frequency of Price Change GCC.do	text/plain	6.5 KB	02/23/2022 08:38:PM

Project Citation:

Xu, Mingzhi, Feenstra, Robert C, and Antoniades, Alexis. Using the Retail Distribution of Sellers to Impute Expenditure Shares. Nashville, TN: American Economic Association [publisher], 2022. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2022-07-27. https://doi.org/10.3886/E163361V1

Project Description

Summary: We provide the code needed to generate the results in "Using the Retail Distribution of Sellers to Impute Expenditure Shares".

Abstract:
Many price indices must be constructed without item-specific quantity data for the elementary expenditure aggregates. The paper shows that for some consumer goods in the Gulf Cooperation Council (GCC) countries and in the United States from 2006 to 2011, one can approximate expenditure shares using weights derived from the retail distribution of sellers.These weights are the share of stores selling an item, or the share of stores adjusted by the total number of items sold in those stores. Relative to using no weights, the paper finds that using retail-distribution-imputed weights substantially reduce bias in the frequency of price changes, in annual inflation, and in price comparisons across countries.

Scope of Project

Subject Terms: scanner data; online price data; retail distribution]; measurement error; code

JEL Classification:
      C81 Methodology for Collecting, Estimating, and Organizing Microeconomic Data; Data Access
      E01 Measurement and Data on National Income and Product Accounts and Wealth; Environmental Accounts
      E31 Price Level; Inflation; Deflation
      E37 Prices, Business Fluctuations, and Cycles: Forecasting and Simulation: Models and Applications

Methodology

Data Source: We use US Nielsen Scanner data for the period 2006 to 2011 for the baseline results. The data are provided through the NielsenIQ Datasets at the Kilts Center for Marketing Data Center at The University of Chicago.

The code provided enables those with access to the data to reproduce our results.

For robustness, we also use Nielsen data from the six GCC countries, also for the period 2006-2011. These data are proprietary and we are not authorized to share based on the agreement with Nielsen.

Finally, for Figure 1 we use Chinese data that come from Nielsen and from on online scraping exercise. These data is also proprietary.

Related Publications

Download this project

Published Versions

V1 [2022-07-27]

Export Metadata

Dublin Core

DDI 2.5

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.