Skip to content

Producer Price Index Standard Errors 2014 This product is designated as National Statistics

Released: 17 June 2014 Download PDF

Abstract

The Producer Price Index (PPI), produced by the Office for National Statistics (ONS), is a monthly survey that measures the price changes of goods bought and sold by UK manufacturers. The output PPI measures changes in the price of goods produced by UK manufacturers. The output price (also known as the factory gate price) reflects the total price received by the manufacturer for a particular product. It includes costs such as labour, raw materials and energy, as well as interest on loans, site and building maintenance or rent. This article presents the calculated standard errors of the output PPI during the period February 2013 to January 2014, for both month-on-month and 12 month growth. These are calculated for all manufacturing (Gross Sector Output (GSO) including duty) and for Standard Industrial Classification (SIC) divisions. Also included in this article are the all manufacturing (GSO including duty) index values and corresponding standard errors for each month over the period February 2013 to January 2014. This has also been calculated at the divisional level for selected divisions, which were chosen either because of their high weight into the GSO or because they were found to have a large standard error.

Standard Errors

If the PPI was based on a census of all businesses in the UK and they all returned correct data, then the true values of price movements would be known. However, since the PPI is based on a sample of UK businesses, it means that population price movements are estimates and are subject to sampling variability. Sampling variability means that if a different set of businesses had been sampled a different estimate would have resulted. Exactly how much these would differ by taking different samples cannot be measured directly, but is approximated using a standard error (calculated as the square root of the variance of the estimate). Standard errors are used to show the spread of price movements and are one way of assessing the accuracy of a price index. The lower the standard error for a given price index, the more confident one can be that the estimated index is close to the true value for the price index had all UK businesses been included.

Month-on-month growth

Table 1 presents the month-on-month growth rates along with standard errors for the all manufacturing PPI (GSO including duty) and the divisional level PPIs. The standard errors have been calculated for each division by averaging each month-on-month growth standard error over the period February 2013 to January 2014, creating a 12 month average.

Table 1: Month-on-month growth rates; by division with 95% confidence intervals February 2013 to January 2014, 12 month average

Two Digit (division) Division Description Month-on-month growth, January 2014 (percentages) Standard error of the month-on-month growth, 12 month average, February 2013 to January 2014 (percentage points)
All Manufacturing   0.0 0.3
10 Food products 0.3 0.2
11 Beverages -0.4 0.2
12 Tobacco 0.0 0.1
13 Textiles 0.4 0.2
14 Wearing apparel 2.3 0.1
15 Leather and related products -0.2 0.3
16 Wood, products of wood and cork, except furniture; articles of straw and plaiting materials 0.4 0.2
17 Paper and paper products 0.2 0.1
18 Printing and recording services, printed matter 0.1 0.3
19 Coke and refined petroleum products -2.8 1.2
20 Chemicals and chemical products 0.5 0.3
21 Basic pharmaceutical products and preparations 0.6 0.4
22 Rubber and plastic products 0.2 0.2
23 Other non-metallic mineral products 0.6 0.2
24 Basic metals -0.1 0.4
25 Fabricated metal products, except machinery and equipment 0.1 0.1
26 Computer, electronic and optical products -0.1 0.3
27 Electrical equipment 0.4 0.1
28 Machinery and equipment NEC 0.6 0.1
29 Motor vehicles, trailers and semi trailers 0.1 0.1
30 Other transport equipment 0.7 0.3
31 Furniture 0.5 0.1
32 Other manufactured goods NEC 0.4 0.2
33 Repair and installation services of machinery and equipment 2.5 0.3

Table source: Office for National Statistics

Table notes:

  1. Division 19 ‘Coke and refined petroleum products’ is based on an 11 month average as there was not sufficient data during October 2013 to create a standard error. Imputed prices were used in place of actual price data in the calculation of index values.

Download table

The median of the divisional standard errors of the month-on-month growths is 0.2. Divisions 21 ‘Basic pharmaceutical products and preparations’ and 24 ‘Basic metals’ have slightly higher standard errors at 0.4 each, due to variation of price movements throughout the period February 2013 to January 2014. Division 19 ‘Coke and refined petroleum products’ has a very high standard error of 1.2, likely to be a result of the volatile nature of price movements within this index.

Although the coverage of businesses who produce coke and refined petroleum products is high compared other SIC divisions (we estimate this to be around 70%) it is the large amount of spread in the price movements provided by these businesses that causes a large standard error.

Figure 1 shows the month-on-month growth rates in January 2014 for all manufacturing PPI (GSO including duty) and divisional level PPIs along with approximate 95% confidence intervals of ± 2 standard errors. This graph highlights the large standard error of division 19 when compared to the standard errors of the other divisions.

Figure 1: Month-on-month growth rates by SIC division, with 95% confidence intervals, February 2013 to January 2014, month average

Figure 1: Month-on-month growth rates by SIC division, with 95% confidence intervals, February 2013 to January 2014, month average
Source: Office for National Statistics

Download chart

12 month growth

Table 2 presents 12 month growth rates along with standard errors for the all manufacturing PPI (GSO including duty) and the divisional level PPIs. The standard errors have been calculated for each division by creating an average of individual 12 month growth standard errors over certain periods of 12 months between February 2013 and January 2014. The number of periods the standard errors have been averaged over depends on the data available.

Table 2: 12 month growth rates with 95% confidence intervals, February 2013 to January 2014

Two Digit (division) Division Description 12 month growth, January 2014 (percentages) Standard error of the 12 month growth, averaged, February 2013 to January 2014 (percentage points) Number of 12 month standard errors used to calculate average
All Manufacturing   0.3 1.2 12
10 Food products 1.7 0.9 1
11 Beverages 1.2 0.7 1
12 Tobacco 8.9 1.1 12
13 Textiles 1.1 0.8 12
14 Wearing apparel 2.7 0.8 12
15 Leather and related products 1.8 2.0 12
16 Wood, products of wood and cork, except furniture; articles of straw and plaiting materials 2.1 0.6 12
17 Paper and paper products 1.6 1.0 12
18 Printing and recording services, printed matter 0.2 1.2 12
19 Coke and refined petroleum products -5.5 5.4 8
20 Chemicals and chemical products -2.7 0.9 12
21 Basic pharmaceutical products and preparations 0.6 1.4 12
22 Rubber and plastic products 0.2 1.3 12
23 Other non-metallic mineral products 1.2 0.7 12
24 Basic metals -2.9 1.3 12
25 Fabricated metal products, except machinery and equipment 0.6 0.4 12
26 Computer, electronic and optical products 2.6 1.4 12
27 Electrical equipment -0.2 0.5 12
28 Machinery and equipment NEC 2.6 0.4 12
29 Motor vehicles, trailers and semi trailers 0.0 0.6 12
30 Other transport equipment 2.2 1.3 12
31 Furniture 1.0 0.4 5
32 Other manufactured goods NEC 1.4 0.7 12
33 Repair and installation services of machinery and equipment 3.0 1.1 12

Table source: Office for National Statistics

Table notes:

  1. Not all divisions have been calculated using a 12 month average due to missing data. The number of months that the standard error has been averaged over is indicated in the column ‘Number of 12 month standard errors used to calculate average’. Imputed prices were used in the calculation of index values.

Download table

The median of the divisional standard errors of the 12 month growths is 0.9. As was the same with the standard errors for the month-on-month growth, Divisions 21 ‘Basic pharmaceutical products and preparations’ and 24 ‘Basic metals’ have higher than average 12 month growth standard errors at 1.4 and 1.3 respectively. Divisions 15 ‘Leather and related products’ and 30 ‘Other transport equipment’ also have higher than average 12 month growth standard errors at 2.0 and 1.3, despite their month-on-month growth standard errors being close to the median. Again division 19 ‘Coke and refined petroleum products’ has the highest standard error at 5.4. As for the month-on-month growth, this is a result of the variation in price movements seen between the businesses sampled and included in the index.

Figure 2 shows the 12 month growth rates in January 2014 for all manufacturing PPI (GSO including duty) and divisional level PPIs along with approximate 95% confidence intervals of ± 2 standard errors. The graph highlights the larger standard errors for divisions 15, 19, 21, 24 and 30. The standard errors of the 12 month growth rates are overall larger than the standard errors of the month-on-month growth rates. This would be expected as price movements would vary more over a 12 month period.


Figure 2: 12 month growth rates by SIC division with 95% confidence intervals, February 2013 to January 2014

Figure 2: 12 month growth rates by SIC division with 95% confidence intervals, February 2013 to January 2014
Source: Office for National Statistics

Download chart

Index levels

Figure 3 shows the all manufacturing (GSO including duty) index values along with 95% confidence intervals of ±2 standard errors for the period February 2013 to January 2014.

Figure 3: All Manufacturing (GSO including duty) index values with 95% confidence intervals, February 2013 to January 2014

Figure 3: All Manufacturing (GSO including duty) index values with 95% confidence intervals, February 2013 to January 2014
Source: Office for National Statistics

Notes:

  1. The October 2013 index value does not have a 95% confidence interval. This is due to a standard error not being calculated for division 19 ‘Coke and refined petroleum products’ because of missing data in this month. Imputed prices were used in the calculation of the index value. The graph shows that the confidence intervals are consistent throughout.

Download chart

The graph shows that the all manufacturing index value standard errors are consistent throughout the period February 2013 to January 2014.

Standard errors were also calculated for index values in divisions which had either a large weight into the GSO or for which the calculated standard error was particularly large.
Figure 4 shows division 10 ‘Food products’ index values along with 95% confidence intervals of ±2 standard errors for the period February 2013 to January 2014.

Figure 4: Food products (SIC division 10) index values with 95% confidence intervals, February 2013 to January 2014

Figure 4: Food products (SIC division 10) index values with 95% confidence intervals, February 2013 to January 2014
Source: Office for National Statistics

Download chart


This division has a large contribution into the top level all manufacturing PPI (GSO including duty) with a weight of 21.3%. While there is some variability of index values, the confidence intervals remain consistent throughout.

Figure 5 shows division 15 ‘Leather and related products’ index values along with 95% confidence intervals of ±2 standard errors for the period February 2013 to January 2014.

Figure 5: Leather and related products (SIC division 15) index values with 95% confidence intervals, February 2013 to January 2014

Figure 5: Leather and related products (SIC division 15) index values with 95% confidence intervals, February 2013 to January 2014
Source: Office for National Statistics

Download chart


This division has a large 12 month growth standard error of 2.0. The graph shows that when the index value increases in May 2013 the confidence interval is wider. The confidence intervals remain large with a peak in January 2014.

Figure 6 shows division 19 ‘Coke and refined petroleum products’ index values along with 95% confidence intervals of ±2 standard errors for the period February 2013 to January 2014.

Figure 6: Coke and refined petroleum products (SIC division 19) index values with 95% confidence intervals, February 2013 to January 2014

Figure 6: Coke and refined petroleum products (SIC division 19) index values with 95% confidence intervals, February 2013 to January 2014
Source: Office for National Statistics

Notes:

  1. The October 2013 index value does not have a 95% confidence interval. This is due to a standard error not being calculated because of missing data in this month. Imputed prices were used in the calculation of the index value.

Download chart


This division has a large contribution into the top level all manufacturing PPI (GSO including duty) with a weight of 11.3%. It also had the largest month-on-month growth standard errors and 12 month growth standard errors at 1.2 and 5.4 respectively. The graph shows wide confidence intervals, due to the volatile movement of prices within this division. However, they remain consistent throughout despite the fluctuating index values.

Figure 7 shows division 24 ‘Basic metals’ index values along with 95% confidence intervals of ±2 standard errors for the period February 2013 to January 2014.

Figure 7: Basic metals (SIC division 24) index values with 95% confidence intervals, February 2013 to January 2014

Figure 7: Basic metals (SIC division 24) index values with 95% confidence intervals, February 2013 to January 2014
Source: Office for National Statistics

Download chart


This division has a large contribution into the top level all manufacturing PPI (GSO including duty) with a weight of 10.3%. It also has a month-on-month growth standard error of 0.4 which is slightly higher than the median of the divisional month-on-month growth standard errors of 0.2. The confidence interval of the index values remains consistent throughout the period.

Figure 8 shows division 30 ‘Other transport equipment’ index values along with 95% confidence intervals of ±2 standard errors for the period February 2013 to January 2014.

Figure 8: Other transport equipment (SIC division 30) index values with 95% confidence intervals, February 2013 to January 2014

Figure 8: Other transport equipment (SIC division 30) index values with 95% confidence intervals, February 2013 to January 2014
Source: Office for National Statistics

Download chart


This division has a large contribution into the top level all manufacturing (GSO including duty) index with a weight of 6.3%. It also has a 12 month growth standard error of 1.3 which is higher than the median of the divisional 12 month growth standard errors of 0.9. The confidence intervals can be seen to gradually increase throughout the period.

Conclusions

Standard errors were found to be broadly similar for all SIC divisions that are aggregated to form the GSO index for all manufacturing, with a few exceptions.

Division 19 ‘Coke and refined petroleum products’ had a significantly higher standard error when compared to the other divisions, due to the volatile price movements within this division. However coverage of businesses included in this division is high, which should minimise the impact of this volatility. The standard errors show consistency in terms of magnitude, which shows that this level of volatility is fairly constant.

Other divisions with significantly high standard errors include division 15 ‘Leather and related products’, division 21 ‘Basic pharmaceutical products and preparations’ and division 26 ‘Computer, electronic and optical products’. This is also be due to volatile price movements within these divisions.

Where standard errors were calculated for index values rather than growths, it was found that most of the standard errors stayed consistent each month throughout the period February 2013 to January 2014 regardless of the size of the standard error.

Background notes

  1. Standard errors have been published previously for PPI in two articles: Measuring the quality of producer price index; and Measuring the quality of producer price index – an update.
  2. The producer price index is published monthly in a Statistical Bulletin.
  3. A Quality and Methodology (QMI) paper for the PPI describes the intended uses of the statistics, their general quality and the methods used to produce them.
  4. The methods used to calculate standard errors presented here can be found in Survey Methodology Bulletin No.62, pages 62-80.
  5. Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: media.relations@ons.gsi.gov.uk

Supporting information

Further information

Producer Price Index - Input and output index series. Contains producer price indices of materials and fuels purchased and output of manufacturing industry by broad sector.
Producer Prices Indices - Price indices on UK manufactured products together with the materials and fuel purchased, and the home sales at both broad and detailed industry levels. The data are used throughout business and government.
Aerospace and Electronic Cost Indices - Cost indices (purchase of materials and fuels, earnings and National Insurance, general expenses and combined costs) relating to four aerospace and electronics industries. The data are based on the revised Standard Industrial Classification SIC (2007) and are calculated on a base year of 2010=100. The indices are widely used by government and business as an authoritative source of information on inflation, cost adjusting and price variation in trading contracts.
Impact of Rebasing the Producer Price Index to 2010=100 - This article outlines the rebasing to the Producer Price Index (PPI) (including trade prices) from 2005=100 to 2010=100.
Content from the Office for National Statistics.
© Crown Copyright applies unless otherwise stated.