This Economic Commentary has presented a new indicator of economic activity. Using principal component analysis, eleven time series from the Riksbank's Business Survey have been weighted together into one indicator. The information summarised in the indicator provides timely signals on the state of the economy, which has been tested by means of correlation analysis and a forecasting exercise using an autoregressive model. The Business Survey indicator has also been compared with other survey-based indicators of economic activity. The results of the correlation analysis and the forecasting exercise for the different indicators do not differ significantly from each other. They are highly correlated (see Table 5 in the Appendix) and generally provide a similar picture of economic activity. However, there are some differences worth highlighting. The Business Survey indicator has both a higher correlation and a smaller forecast error than the NIER's BS confidence indicator – the indicator that most closely resembles the Business Survey indicator in terms of purpose and questions included. The NIER's Economic Tendency Survey has a higher correlation with GDP and smaller forecast errors for real-time data, but the Business Survey indicator has slightly smaller forecast errors for the most recently published GDP data. How the Business Survey indicator relates to the Economic Tendency Survey is particularly interesting as the Economic Tendency Survey is based on survey data from both companies and households. Although the Riksbank's Business Survey captures household behaviour and attitudes indirectly by asking companies about sales, it lacks the direct link to households. The relationship between the Economic Tendency Survey and the Business Survey indicator is strong with a correlation coefficient of 0.89. This reinforces the interpretation that the Business Survey indicator is a reliable measure of economic activity.

As the quantitative responses summarised in the indicator are consistent with the qualitative responses provided by the companies, the results of the analysis indicate that they are also reliable. In summary, this commentary has shown that the data collected by the Business Survey are a valuable complement to other economic activity data and that the Business Survey indicator presents these data effectively. In the quarters in which the business survey is conducted, the Business Survey indicator is published one to two months before the GDP data for the period in question, which means that it can provide an early picture of the state of the economy.