Skip to main content


DataBasics walks you through the essentials of economic data manipulations. The articles present numeric operations that economists use to make data more meaningful. The data definitions are descriptions of frequently used Texas economic variables that, taken together, help paint a picture of statewide economic activity.

Dive into the articles

Indexing numerical data to a common starting point is useful in a variety of business, financial and economic contexts. It allows for quick comparison and ranking. For example, equity traders index stock prices and stock indexes to compare performance over time, and economists index data to prominent events—say economic peaks (or troughs)—to see how the data decline or rise relative to each other. Indexed data are useful because they allow an observer to quickly determine rates of growth by looking at a chart's vertical axis. They also allow for comparison of variables with different magnitudes, such as state and U.S. employment data.

Annualized growth rates reflect the amount a monthly or quarterly variable would have changed over a year's time, had it continued to grow at the given rate. The result is a percent change that facilitates comparison of growth rates of differing time lengths. For instance, economists annualize quarterly percent changes in gross domestic product (GDP) so that one-quarter growth comparisons can be made to the annual growth rate in previous years.

Smoothing techniques are designed to reduce or eliminate short-term volatility in data series, which allow researchers to better decipher economic trends. One simple type of smoothing is to take a moving average of the series’ values, which is often centered on the middle month so as not to change the timing of the series.

Many economic data series, such as GDP and exports are adjusted for a rise in the price level, or inflation. Because $1 will buy less than it did 20 years ago, one needs to take account of inflation to understand the real purchasing power of that dollar. A simple methodology can be used to “deflate” a current dollar, or nominal, data series such as retail sales, to inflation-adjusted, real values using a common price index such as the CPI (consumer price index).

When comparing economic data series, researchers look at absolute levels as well as growth rates. Levels are useful in analyzing differences in similar entities like per capita income in neighboring states, or output (GDP) levels among countries.

Growth rates, or percentage changes, are used more often when comparing data over time periods, for instance, looking at a region’s job growth in the last year versus the prior year. Growth rates also allow for better comparison of economic performance across regions of differing size. For example, comparing the level change of GDP in Texas to that of Rhode Island would not be that useful given the large size differences in the two economies, but looking at their growth rates would give important insights.

Many data series such as employment, home sales and retail sales exhibit seasonal patterns—that is, they have large predictable changes around the same time of year. For example, the summer school break or improved weather in the spring can affect retail sales, showing an increase every year. While seasonal patterns are important, to understand the underlying growth in the data it is often useful to seasonally adjust the data and thus remove the normal seasonal changes. In seasonally adjusting data, the process formally estimates the normal growth or decline for each of the months or quarters of the year and subtracts these movements from the series. To make sure the annual growth in the series is not altered the adjustments over any 12-month or four-quarter period sum to zero.

As described in Seasonally Adjusting Data, many data series change in the same direction and magnitude at certain times of the year. For data series that are constructed from different sources, standard seasonal adjustment may create problems if the magnitudes of the seasonal changes differ across the data sources. Most notably, this arises with payroll employment data. The Dallas Fed's two-step seasonal adjustment involves seasonally adjusting each source of data separately, and then combining the sources to create a continuous employment series.

Most data series are revised after their initial release as new information becomes available. Early benchmarking refers to the Dallas Fed's quarterly revision of payroll employment estimates for Texas. By incorporating administrative data from the Texas Workforce Commission, early benchmarking ensures that the Dallas Fed's payroll employment estimates use the most comprehensive information available.