Annual analysis of revisions to money, credit and effective interest rates data (2014-16)

Each year the Bank publishes analysis of the revisions to its monthly data on money, credit, and effective interest rates produced by the Data and Statistics Division.
Published on 29 July 2019

By Angel Chu, Sienna Holcombe and Jessica Verlander

Each year the Bank publishes analysis of the revisions to its monthly data on money, credit, and effective interest rates produced by the Data and Statistics Division (DSD). This year’s analysis shows that the revisions can be considered as immaterial for most series tested. This is the same broad conclusion as the 2018 analysis of 2013-15 data and the 2017 analysis of 2012-14 data.

Revisions are a normal part of the data production process. There are several reasons why data might be revised after its initial publication. Reporters of data to the Bank may submit corrections to earlier data if they discover errors or make improvements to their data systems. In addition, the Bank might change the methodology it uses to produce the data. Also, the seasonal adjustment process can lead to revisions to an entire series, as each new data point provides new information about the seasonal pattern of the data.

Revisions analysis gives users an indication of how much weight to place on data when it is first released. Early releases of data that tend to receive few revisions can be regarded as less noisy and more reliable.

Revisions can be measured in different ways. The tables below show a variety of metrics for the money and credit and effective rates data. Tables 1 and 2 illustrate the size of revisions, showing the mean revision, mean absolute revision and average published outturn. Tables 3 and 4 present metrics to analyse bias in the data. For each series, a T-test is conducted. This is the simplest test for bias, and is a function of the mean and standard deviation of the revisions, and the number of observations. However, this test is only properly valid when the sample of revisions is independently and identically distributed, and fails if revisions are subject to autocorrelation or to non-constant variance (heteroskedasticity). Tables 3 and 4 include alternative tests for bias that control for this. The adjusted t-test takes into account any first order autocorrelation in the revisions; and the Newey-West test is valid when heteroskedasticity and autocorrelation is present in revisions to the previous two months of data. Tables 5 and 6 provide further analysis of the revisions, showing the root mean squared revision, the ratio of this to the mean of the revised data, and ratio of the mean squared revision to the variance of the revised data.

To view all related tables, please download the full article:

PDF Annual analysis of revisions to money, credit and effective interest rates data (2014-16)

For questions relating to this article please contact dsd_ms@bankofengland.co.uk or call +44 (0) 20 3461 5356.