Statistics for Decision Making in Economics

Statistics for Decision Making in Economics

Introduction

The decision making process that virtually all of our federal agencies use in the determination of our economic direction are based upon descriptive and inferential statistical data gathered from our economy. The forecasts and predictions derived from the resulting data and statistical models have the potential of affecting the welfare of the country as a whole and all the people within it. Therefore the sampling and analysis of the source data once it has been gathered is extremely important. Because of the sheer number of statistics that are collected, no one agency could hope to gather and analyze all the sample data involved. Hundreds of federal, state, and private agencies are utilized to gather the sample data and analyze the various statistics which larger federal used for the direction of our economy. For purposes of depth, this paper will concentrate on the statistics generated by the Bureau of Economic Analysis, the Federal Reserve, and the Bureau of Labor Statistics. Together these agencies generate a set of economic accounts called the U.S. national economic accounts, which provides key indicators of growth, income, and expenditures of the U.S. and are utilized to define the direction of monetary policy within the U.S. by the Federal Reserve.

Bureau of Economic Analysis

The Bureau of Economic Analysis ( BEA ) is an agency of the Economics and Statistics Administration which is part of the Department of Commerce. The ( BEA ) generates a group of statistics call the US national income and product accounts ( NIPAs ). These are a set of point estimate economic accounts which detail US national output and the distribution of income from US production. The NIPAs are "closely watched" (Bureau of Economic Analysis, 2010, p 6 ) economic statistics, which influence decisions by many government agencies, industries, and individuals. In concert with their own and other time-series data, they are used for forecasting models and trends in the economy. The most important of these statistics is the Global Domestic Product ( GDP ), which is the most recognized measure of US production.

Sources of Data.

The descriptive statistics that the Bureau of Economic Analysis collects are sourced from samples gathered by a number of other federal, state, and private agencies. The majority of data comes from federal agencies such as the US Census Bureau, the Bureau of Labor and the Treasury Department. Additional sample data is gathered from the data of industries and other private sources.

Multiple data sources are cross-checked against each other to insure accuracy and eliminate any error of estimation. The finalized analysis numbers are cross-checked against separate analysis to further ensure reduced error. For instance the GDP is known to equal to GDI and also the Gross Value Added.

Statistical Methodologies

A variety of estimating methodologies are used in the preparation of NIPA accounts each method used is in accordance with achieving the best estimate accuracy. These methods are changed with improvements in source data and methodologies and are evaluated to ensure they provide the most reliable estimate and understanding of the economy. The actual methodology that the BEA uses are "complex statistical techniques that are designed to ensure consistency across the entire time series for a given estimate and between interrelated estimates" (Bureau of Economic Analysis, 2010, p 58 ) These methodologies can be generalized into the following :

BEA adjusts the source data to conform to NIPA definitions, augments the primary data source from other sources, and values all goods at the port value in order to provide a level way of judging manufacturer good values as part of their estimates.

Seasonal adjustments are made to the data when it indicates significant seasonal patterns. ( i.e. the effects of holidays or weather )

Data smoothing is implemented using moving averages on shorter time-series calculations.

Estimates are sometimes generated using interpolation and extrapolation methods in cases where short term time-series data is not as reliable as annual data. Currently the BEA is using the following quadratic minimization, for interpolation :

min┬(x_t )⁡∑_(t=2)^4N▒〖〖(x_t/z_t -x_(t-1)/z_(t-1) )〗^2, s.t.∑_(t=1)^4▒〖x_t=A_(1,... ) 〗 ∑_(t=4N-3)^N▒〖X_t= A_N 〗〗

"where z is the indicator series, x is the interpolated series, A are the annual controls that the interpolated series must sum to, and N is the number of years for the interpolation." (Bureau of Economic Analysis, 2010, p 63)

Estimates for NIPA aggregates are estimated using a fisher formula that incorporates weights for 2 adjacent years. The formula follows :

Q_t^F= √((∑▒〖p_(t-1) q_t 〗)/(∑▒〖p_(t-1) q_(t-1) 〗) X (∑▒〖p_t q_t 〗)/(∑▒〖p_t q_(t-1) 〗))

"where the p's and q's represent prices and quantities of detailed components in the 2 years." (Bureau of Economic Analysis, 2010, p 69)

The Federal Reserve

The Federal Reserve System (FED) in the central bank of the United States. The Board of Governors acts as directing body for the Federal Reserve and its branch banks. The FED determines monetary policy and enacts this policy through control of the federal interest rate and regulation of the money supply. As an independent central bank the FED has been set up to operate in such a way as to avoid political interference in its operations.

The Federal Reserve has a Research and Statistics Division dedicated solely to the analysis of economic and financial data for the use of the Federal Reserve Board and other Federal Reserve officials. The forecasting models and trend analysis developed by this division act as the foundation for the formulation of monetary, regulatory, and supervisory policies. The division additionally releases this information to the public for further analysis and is responsible for helping understanding of economic policy.

Sources of Data

The FED collects statistical sample data from the depository institutions with which it does business, other federal agencies, and from its own operations as the central bank.

Statistical Methodologies

Not much is published on the exact methods used by the Federal Reserve

...