What is seasonally adjusted economic data?

Summary

  • Seasonal adjustment is an integral tool used to map trends in the economy that are occurring on a non-seasonal and a non-cyclical basis.
  • Seasonal effects are one of the three components of time series including trends and irregularities.
  • Seasonal effects give a misguided projection of the inherent trend in the demand and supply patterns and are consequently eliminated from raw data.

‘Seasonally adjusted data’ is a term often mentioned by economists and statistical organisations while presenting important economic figures. However, it is essential to understand what the term means and the mechanism behind.  

Seasonal adjustment is nothing but the elimination of cyclical trends observed in a time series data. For the uninitiated, this statement would make little to no sense. Let us take a closer look at how raw data is made seasonally adjusted data and why this process is carried out.

ALSO READ: How is inflation calculated?

Time Series Data

Time series data refers to a set of data recorded at equal intervals. Time series data has data points corresponding to each time entry. Most economic data collected for research is collected as a time series data only. This includes quarterly rates of CPI, Monthly data on retail sales, yearly data on unemployment, etc.

All the above-mentioned indicators are measured at a specific time and are represented as time series. This helps establish a trend observed in the data over time and helps draw a comparison between different equally spaced periods. This leads to the data absorbing cyclical trends observed over a long period.

Apart from the time series arrangement, data can also be arranged in a cross-sectional manner or as panel data. Cross-sectional data includes information that is collected at a fixed point in time and is arranged based on different variables. Panel data is a mix of both cross-sectional data and time series data.

RELATED: What is Unemployment?

Components of Time Series

Time series data comprises of three different types of casual factors that lead to changes in the data. These include:

  • Seasonal Effects: This includes seasonal trends and calendar-related trends which occur in a cycle and are observed repeatedly, year after year. There are various reasons why repetitive patterns are observed in data, including the changing weather which includes increased electricity consumption during winters, cultural events which influence buying patterns of consumers and administrative measures like opening of schools after summer break. Calendar-related changes are also essential, including moving holidays and the varying number of weekends in a month.
  • Trend: The trend observed in data is what analysts are interested in. Trends represent the change in the behaviour of the variable being observed and shed light on how non-seasonal factors have contributed to the change. Non-seasonal factors are simply micro or macro-level changes that do not occur at a specific time or at regular intervals. The trend component is integral in capturing the business cycle prevalent in the economy.
  • Shocks or Irregularities in Data: These unprecedented effects can cause the data to show high variation. For instance, a statistical error in sales data can wrongly project consumer behaviour without an explicable reason. Additionally, shocks to the economy like a natural calamity or a pandemic can lead to sudden turns in the data.

RELATED: What is Behavioural Economics?

Seasonal Adjustment and its Requirement

There are certain seasonal events or factors that affect the time series data. These factors give a misguided projection about the real or actual trends happening in the economy. An example to better understand this is to consider the Black Friday sales that occur in November. This sale projects an unusually larger amount of retail sales in the economy during the month. However, the reason for this unusual uptick in sales is known and this uptick affects the actual data figures which are being analysed.

The intention of observing retail data in time series is to establish and observe trends and to tap the consumer preferences in the sector of interest. This study is altered when seasonal and cyclical trends come into the picture. Thus, it becomes important to eliminate the effect of these trends on the data. Cyclical trends are different from seasonal trends as they may not repeat after every calendar year. Instead, they can recur within time spans shorter or longer than a year.

The data that includes these seasonal trends is known as raw data. However, projections and analysis of trends based on raw data can be inaccurate. Thus, seasonal adjustment is a handy mechanism used by economists and statistical experts to map the changes in economic indicators.

How is Seasonal Adjustment done?

There are intricate statistical models used to remove seasonal trends from raw data. The process entails evening out those upticks and downturns observed in the supply and demand which are known to be a consequence of seasonal factors.

An average is taken for each month/quarter in the data for a whole year. The ratio between the actual number and the average can be used to describe the seasonal factor during that period. This seasonality factor can be used to calculate Seasonally Adjusted Annual Rate, which is a tool used to remove seasonal trends.

INTERESTING READ: How Does Behavioural Economics Influence Market Participants?

Comment


Disclaimer