Showing posts with label statistics. Show all posts
Showing posts with label statistics. Show all posts

Thursday, August 5, 2021

Monday, July 8, 2019

New NBER Working Paper: Improving the accuracy of economic measurement with multiple data sources: The case of payroll employment data

A new NBER Working Paper, "Improving the accuracy of economic measurement with multiple data sources: The case of payroll employment data," by Cajner, Crane, Decker, Hamins-Puertolas, and Kurz.

Abstract:
This paper combines information from two sources of U.S. private payroll employment to increase the accuracy of real-time measurement of the labor market. The sources are the Current Employment Statistics (CES) from BLS and microdata from the payroll processing firm ADP. We briefly describe the ADP-derived data series, compare it to the BLS data, and describe an exercise that benchmarks the data series to an employment census. The CES and the ADP employment data are each derived from roughly equal-sized samples. We argue that combining CES and ADP data series reduces the measurement error inherent in both data sources. In particular, we infer “true” unobserved payroll employment growth using a state-space model and find that the optimal predictor of the unobserved state puts approximately equal weight on theCES and ADP-derived series. Moreover, the estimated state contains information about future readings of payroll employment.
Available at NBER

Saturday, April 15, 2017

FEE: "The CPI is a False Guide for Monetary Policy"

An Austrian economist, Richard M. Ebeling,  takes on the Consumer Price Index.
The pricing subcategories highlight the smoke and mirrors that is the statisticians’ distinction between overall and “core” inflation. People will occasionally enter the market to purchase a new stove, couch, or bedroom set, and if the prices for these goods happen to be going down, or slowly rising, we may sense that our dollar is going further than in the past.

But buying goods like these is an infrequent event for virtually all of us. On the other hand, every one of us are in the marketplace paying for food, gas for our cars, paying heating and electric bills on a regular basis. The prices of these goods and services, in the specific brands and combinations that we as individuals choose to buy, are what we personally experience as a change in the cost-of-living and our personal rate of price inflation (or price deflation).

The Consumer Price Index is an artificial statistical creation, derived from thousands of individual prices, a statistical composite that only exists in the statistician’s calculations. It is the individual goods in the subcategories of goods that determine the change in the cost-of-living and the degree of price inflation (or deflation) that we each experience.
Read the whole article here.

Indicators

Test