The U.S. Census Bureau Index of Economic Activity (IDEA) is an aggregation of 15 of the Census Bureau’s primary economic data series that provides a single time series constructed as a weighted average.
Last Updated 11:30 a.m. November 8th, 2022
The U.S. Census Bureau Index of Economic Activity (IDEA) is an aggregation of 15 of the Census Bureau’s primary economic data series that provides a single time series constructed as a weighted combination. This monthly index is constructed from time series of actual and estimated monthly growth rates. As the underlying data are subject to revision, the IDEA is calculated daily, incorporating the most recent values, new estimates, and revisions in each component series.
Values for the index are available starting at August 2004 and will be updated on each business day. Historical data of the index are available for download on the “Downloads” tab.
Access the IDEA methodology, IDEA FAQs, and IDEA Working Paper for more information.
Feedback can be shared by email at EID.Economic.Index@census.gov.
The IDEA is an aggregation of 15 of the Census Bureau’s primary economic data series that provides a single time series constructed as a weighted average. This monthly index is constructed from time series of actual and estimated monthly growth rates of these 15 series.
Series
Source
Business Applications
Exports of Goods and Services
Housing Units Authorized in Permit-Issuing Places
Housing Units Completed
Housing Units Started
Imports of Goods and Services
Manufacturing Inventories
Manufacturing Value of New Orders
New Orders for Durable Goods
New Single-Family Houses for Sale
New Single-Family Houses Sold
Retail Trade and Food Services Sales
Retail Inventories
Total Construction Spending
Wholesale Trade Inventories
The IDEA will be updated each business day to ensure it reflects each new and/or updated published value(s) of the input data series. There will be some days where no input series change, and the index value will remain the same as it was on the previous day.
Each business day, two datasets will be updated. All daily vintages of the below datasets will be available for download.
- Monthly IDEA values back to August 2004.
- IDEA inputs of published economic data series values.
In addition, the dataset of nowcasted values* for economic data series that were used to compute the index before they were published for the current month will be available for download at the end of the data processing month.
The principal component weights* for each of the input economic data series are updated annually, beginning August 14th, 2024. The dataset of current and historic weights is available for download.
*For more information on principal components and nowcasting, see the IDEA Methodology.
No. The IDEA uses only publicly available data in its construction. See table in FAQ #2 for the list of included data series.
The IDEA is found at the top of the U.S. Census Bureau Briefing Room webpage,
https://www.census.gov/economic-indicators/.
Datasets can be accessed on the "Downloads" tab within the visualization on this webpage.
We continue to explore adding other relevant economic data series to the IDEA in the future. We are also researching how to incorporate other series that are not on a monthly frequency, like the Quarterly Services Survey.
With respect to the IDEA, the term “growth rate” refers to the log(current month value) – log(prior month value). 100 times the growth rate is approximately equal to the month-to-month
percentage change when the latter is less than 20 percent in magnitude. This approximation deteriorates for percentage changes of larger magnitude.
The month-to-month percentage changes for each monthly economic data series included on the IDEA plots match the values provided by each economic data product in the Briefing Room
(https://www.census.gov/economic-indicators/).
Yes, all input data series are seasonally adjusted. Read more about Census seasonal adjustment at https://www.census.gov/topics/research/seasonal-adjustment.html.
The IDEA time series is presented in a process control chart. Control charts are useful because they help distinguish between variation that is inherent in a process and is therefore expected
versus variation that is unusual and, in many cases, unexpected. The IDEA combines measures from fifteen distinct monthly time series into a single series that is then standardized to have a
mean of zero and a standard deviation of 1. Consequently, the centerline of the control chart is at zero. The chart highlights four distinct regions: (1) values falling within ±1 standard
deviation of the centerline (darkest shade of blue in Figure 1 below), (2) values between ±2 standard deviations and ±1 standard deviation of the centerline (third lightest shade of
blue in Figure 1), (3) values between ±3 standard deviations and ±2 standard deviations of the centerline (second lightest shade of blue in Figure 1), and (4) values outside ±3
standard deviations of the centerline (lightest shade of blue in Figure 1).
All processes – including the IDEA – have some form of variation. Regardless of how well a production process operates, there will always be an inherent or natural variability that exists.
This natural variability or “background noise” is considered normal while the data points fall within determined limits or do not exhibit a trend. In the framework of statistical quality
control, this variability is often referred to as “a stable system of chance causes.” A process that is operating with only chance causes of variation present is said to be in statistical
control.
Few processes remain stable forever. The IDEA process could, for example, be affected by an economy-wide event, or one of the component series might be affected by a major weather event.
Sources of variability that are not part of the chance causes of variation are called “assignable causes” or “special causes.”
In monitoring the IDEA control chart, we recommend considering the following indicators to assess stability (Tague 2004) using the bands in various shades of blue as a guide:
- A single point that is outside the control limits (lightest shade of blue);
- Four out of five consecutive points on the same side of the centerline and exceeding 2 standard deviations (second lightest shade of blue);
- Two out of three consecutive points on the same side of the centerline and exceeding 1 standard deviation (third lightest shade of blue);
- Eight consecutive points that are on the same side of the centerline (in any of the blue bands);
- An obvious consistent, nonrandom pattern (e.g., an upward or downward monotone trend).
Tague, N.R. (2004). The Quality Toolbox (2nd Edition). ASQ Quality Press.
A zero value of the IDEA indicates that the current value is not different from IDEA’s long-term mean (computed using the index values from August 2004 through the most recent May, but excluding the values from March 2020 through August 2020).
As is standard for principal components analysis, the PCA weights are scaled so the sums of the squares of the PCA weights equals 1.
The Main screen highlights changes to the Current Month’s index value. At the start of a new data month, there will be no previous value as it is the first day the Current Month is available. To see history of the Index Value for any particular month, hover over the Index line in the visualization or download the annual “Index_Values.csv” files on the “Downloads” tab.
Yes, effective August 14th, 2024, the span of data used to compute the index was updated. Originally, the index was computed using data from August 2004 through February 2020. The updated span
is August 2004 through the most recent May, excluding the data values from March 2020 through August 2020.
The span determines the data used for both computing the principal component analysis (PCA) weights and for fitting the time series model used for nowcasting. The PCA weights and model fitting
results will now be updated once a year using data through that year's final May input values. See the “Weights.csv” file on the “Downloads” tab for historic PCA weight values. Vintages of the
daily IDEA values before and after this update can be viewed in the annual “Index_Values.csv” files, also available on the “Downloads” tab.
This update uses more recent data in the PCA and model fitting analyses, while fixing the PCA weights and model coefficients through the year. This is done so that changes in the Index values
are driven by new and updated data for the indicator input series and not by changes in the PCA weights or model coefficients. Additional details on this update are included in the IDEA Methodology document.
chevron_left
chevron_right
- Sun
- Mon
- Tue
- Wed
- Thu
- Fri
- Sat