Computer and peripherals price indexes (CPPI)

Detailed information for March 2017

Status:

Active

Frequency:

Monthly

Record number:

5032

The Computer and Peripherals Price Indexes (CPPI) are monthly series measuring changes over time in the price of computers, computer peripherals and smartphones sold to governments, businesses and households.

Data release - April 28, 2017

Description

The Computer and Peripherals Price Indexes (CPPI) are monthly series measuring changes over time in the price of computers, computer peripherals and smartphones sold to governments, businesses and households. The methodology for producing these series employs the hedonic method, the result being an index series that tracks pure price change.

These index series are used by economists, industry analysts and the general public to track and comprehend events and trends in this important contributor to the Information and Communication Technology (ICT) sector.

Within Statistics Canada, the series pertaining to households are used in the calculation of the Consumer Price Index. Several series are used by the Canadian system of macroeconomic accounts in deflating the value of gross investment by the business and government sectors.

Reference period: The time period for which the CPPI equals 100 is the year 2015.

Collection period: The week of the 15th day of every month.

Subjects

  • Information and communications technology
  • Prices and price indexes

Data sources and methodology

Target population

The universe for the CPPI consists of all vendors or distributors of computers, computer peripherals and smartphones in Canada.

Instrument design

This methodology type does not apply to this statistical program.

Sampling

A representative cut off sample of products is chosen, monitored and replenished from the administrative data provided for the following indices:

- Desktop
- Laptop
- Tablets
- Servers
- Smartphones.

For computer monitors and computer printers, however, all of the models that match in two consecutive months are used to estimate the index.

Data sources

Data are collected from other Statistics Canada surveys and/or other sources.

Data are collected by IDC through distributors and retailers websites.

The data is bought from International Data Corporation (IDC) and therefore doesn't fall under the Statistics Act. The data is used to produce price indexes. The data processing conducted by IDC is as follows:
1) Run automatic data collection processes for those sources where data can be collected automatically via web scraping technology and other tools.
2) Manually collect data from remaining (non-automated) sources like distributor and retailer monthly data feeds.
3) A manual review of all raw data collected in phases 1 and 2 is conducted, with a focus on resolving any duplicate SKUs as well as finding and input of any missing data characteristics.
4) Oddities in the data (eg. disproportionate price changes) are automatically flagged for manual review. These oddities are manually reviewed and resolved.
5) The final manual inspection of the deliverable report finds and resolves any remaining duplicate SKUs.

Error detection

Outliers are identified by calculating the price movement of a product from one reference month to the previous month. For large movements, research is done on the price of the model to determine the reason for the movement (for instance, if there was a sale). If justification for the price movement is not found, the data providers (IDC) are asked for reasons. When the price movement is still not justified, the product is excluded from the estimation.

For desktop, laptop, tablet, servers and smart phones, discontinued models are replaced by identifying a model replacement that is as close as possible to the discontinued model by comparing its components. For desktop and laptop computers, hedonic adjustments are applied to the new model to make it as comparable to the old model as possible.

Imputation

Imputation is generally carried out for missing data. In this situation, an estimate is generally made based on the price movement of similar products. Imputation is also done for handling quality change associated with replacements to products. Hedonic coefficients are used to adjust for quality change between discontinued and new models.

Estimation

Weights

The weights used in the various price indexes are based on the shipment information provided by IDC. The shipment information covers the number of units shipped by vendors, as identified by the sub-groups for a particular product. Weights for the computer series, the peripherals series and the smartphone series are updated annually.

Prices

The price and model specification information are supplied by IDC. The prices used are street prices and do not include taxes or shipping and handling charges. One important aspect of these series is the continual and notable improvement in the quality of this category of goods. As a result, the treatment of quality change becomes an important issue. When the pricing for a model in the sample is no longer available, a replacement model is linked in using a hedonic quality adjustment technique to provide an imputed link-in price. This treatment accounts for any substantial differences in quality between the replacement model and the discontinued model.

Quality evaluation

Given that product prices are tracked over time, the reported data are compared to previous cycles to identify any outliers. Any product having a large price change is reviewed by verifying that the product information hasn't changed, doing some internet verifications that the price has in fact changed and to identify any reasons for the price movement. If reasons for the movements are still unknown (this is rarely the case), the data provider is asked to investigate their systems to ensure that there were no data collection issues.

Disclosure control

Statistics Canada is prohibited by law from releasing any data which would divulge information obtained under the Statistics Act that relates to any identifiable person, business or organization without the prior knowledge or the consent in writing of that person, business or organization. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

Individual price and weight data collected under the Statistics Act are confidential. Price data are converted to price indexes and data are released so that it is not possible to identify the suppliers of the raw price information.

Revisions and seasonal adjustment

The index is subject to a six-month revision period and data is not seasonally adjusted.

Data accuracy

The statistical accuracy of this index depends on price and weight data, which are obtained from an outside data source (IDC). Through continuous monitoring of the data received and discussion with the supplier, an acceptable level of data quality and integrity is maintained for the purposes of producing the price index series.

Due to the fact that administrative data is used, confidence intervals are not currently estimated and statements about the statistical reliability cannot be made.

Date modified: