Objectives, uses and users
Concepts, variables and classifications
Coverage and frames
Sampling
Questionnaire design
Response and nonresponse
Data collection and capture operations
Editing
Imputation
Estimation
Revisions and seasonal adjustment
Data quality evaluation
Disclosure control

1. Objectives, uses and users

1.1. Objective

The Monthly Retail Trade Survey (MRTS) provides information on the performance of the retail trade sector on a monthly basis, and when combined with other statistics, represents an important indicator of the state of the Canadian economy.

1.2. Uses

The estimates provide a measure of the health and performance of the retail trade sector. Information collected is used to estimate level and monthly trend for retail sales. At the end of each year, the estimates provide a preliminary look at annual retail sales and performance.

1.3. Users

A variety of organizations, sector associations, and levels of government make use of the information. Retailers rely on the survey results to compare their performance against similar types of businesses, as well as for marketing purposes. Retail associations are able to monitor industry performance and promote their retail industries. Investors can monitor industry growth, which can result in better access to investment capital by retailers. Governments are able to understand the role of retailers in the economy, which aids in the development of policies and tax incentives. As an important industry in the Canadian economy, governments are able to better determine the overall health of the economy through the use of the estimates in the calculation of the nation’s Gross Domestic Product (GDP).

2. Concepts, variables and classifications

2.1. Concepts

The retail trade sector comprises establishments primarily engaged in retailing merchandise, generally without transformation, and rendering services incidental to the sale of merchandise.

The retailing process is the final step in the distribution of merchandise; retailers are therefore organized to sell merchandise in small quantities to the general public. This sector comprises two main types of retailers, that is, store and non-store retailers. The MRTS covers only store retailers. Their main characteristics are described below. Store retailers operate fixed point-of-sale locations, located and designed to attract a high volume of walk-in customers. In general, retail stores have extensive displays of merchandise and use mass-media advertising to attract customers. They typically sell merchandise to the general public for personal or household consumption, but some also serve business and institutional clients. These include establishments such as office supplies stores, computer and software stores, gasoline stations, building material dealers, plumbing supplies stores and electrical supplies stores.

In addition to selling merchandise, some types of store retailers are also engaged in the provision of after-sales services, such as repair and installation. For example, new automobile dealers, electronic and appliance stores and musical instrument and supplies stores often provide repair services, while floor covering stores and window treatment stores often provide installation services. As a general rule, establishments engaged in retailing merchandise and providing after sales services are classified in this sector. Catalogue sales showrooms, gasoline service stations, and mobile home dealers are treated as store retailers.

2.2. Variables

Sales are defined as the sales of all goods purchased for resale, net of returns and discounts. This includes commission revenue and fees earned from selling goods and services on account of others, such as selling lottery tickets, bus tickets, and phone cards. It also includes parts and labour revenue from repair and maintenance; revenue from rental and leasing of goods and equipment; revenues from services, including food services; sales of goods manufactured as a secondary activity; and the proprietor’s withdrawals, at retail, of goods for personal use. Other revenue from rental of real estate, placement fees, operating subsidies, grants, royalties and franchise fees are excluded.

Trading Location is the physical location(s) in which business activity is conducted in each province and territory, and for which sales are credited or recognized in the financial records of the company. For retailers, this would normally be a store.

Constant Dollars: The value of retail trade is measured in two ways; including the effects of price change on sales and net of the effects of price change. The first measure is referred to as retail trade in current dollars and the latter as retail trade in constant dollars. The method of calculating the current dollar estimate is to aggregate the weighted value of sales for all retail outlets. The method of calculating the constant dollar estimate is to first adjust the sales values to a base year, using the Consumer Price Index, and then sum up the resulting values.

2.3. Classification

The Monthly Retail Trade Survey is based on the definition of retail trade under the NAICS (North American Industry Classification System). NAICS is the agreed upon common framework for the production of comparable statistics by the statistical agencies of Canada, Mexico and the United States. The agreement defines the boundaries of twenty sectors. NAICS is based on a production-oriented, or supply based conceptual framework in that establishments are groups into industries according to similarity in production processes used to produce goods and services.

Estimates appear for 21 industries based on special aggregations of the 2007 North American Industry Classification System (NAICS) industries. The 21 industries are further aggregated to 11 sub-sectors.

Geographically, sales estimates are produced for Canada and each province and territory.

3. Coverage and frames

Statistics Canada’s Business Register ( BR) provides the frame for the Monthly Retail Trade Survey. The BR is a structured list of businesses engaged in the production of goods and services in Canada. It is a centrally maintained database containing detailed descriptions of most business entities operating within Canada. The BR includes all incorporated businesses, with or without employees. For unincorporated businesses, the BR includes all employers with businesses, and businesses with no employees with annual sales that have a Goods and Services Tax (GST) or annual revenue that declares individual taxes.  annual sales greater than $30,000 that have a Goods and Services Tax (GST) account (the BR does not include unincorporated businesses with no employees and with annual sales less than $30,000).

The businesses on the BR are represented by a hierarchical structure with four levels, with the statistical enterprise at the top, followed by the statistical company, the statistical establishment and the statistical location. An enterprise can be linked to one or more statistical companies, a statistical company can be linked to one or more statistical establishments, and a statistical establishment to one or more statistical locations.

The target population for the MRTS consists of all statistical establishments on the BR that are classified to the retail sector using the North American Industry Classification System (NAICS) (approximately 200,000 establishments). The NAICS code range for the retail sector is 441100 to 453999. A statistical establishment is the production entity or the smallest grouping of production entities which: produces a homogeneous set of goods or services; does not cross provincial boundaries; and provides data on the value of output, together with the cost of principal intermediate inputs used, along with the cost and quantity of labour used to produce the output. The production entity is the physical unit where the business operations are carried out. It must have a civic address and dedicated labour.

The exclusions to the target population are ancillary establishments (producers of services in support of the activity of producing goods and services for the market of more than one establishment within the enterprise, and serves as a cost centre or a discretionary expense centre for which data on all its costs including labour and depreciation can be reported by the business), future establishments, establishments with a missing or a zero gross business income (GBI) value on the BR and establishments in the following non-covered NAICS:

• 4541 (electronic shopping and mail-order houses)
• 4542 (vending machine operators)
• 45431 (fuel dealers)
• 45439 (other direct selling establishments)

4. Sampling

The MRTS sample consists of 10,000 groups of establishments (clusters) classified to the Retail Trade sector selected from the Statistics Canada Business Register. A cluster of establishments is defined as all establishments belonging to a statistical enterprise that are in the same industrial group and geographical region. The MRTS uses a stratified design with simple random sample selection in each stratum. The stratification is done by industry groups (the mainly, but not only four digit level NAICS), and the geographical regions consisting of the provinces and territories, as well as three provincial sub-regions. We further stratify the population by size.

The size measure is created using a combination of independent survey data and three administrative variables: the annual profiled revenue, the GST sales expressed on an annual basis, and the declared tax revenue (T1 or T2). The size strata consist of one take-all (census), at most, two take-some (partially sampled) strata, and one take-none (non-sampled) stratum. Take-none strata serve to reduce respondent burden by excluding the smaller businesses from the surveyed population. These businesses should represent at most ten percent of total sales. Instead of sending questionnaires to these businesses, the estimates are produced through the use of administrative data.

The sample was allocated optimally in order to reach target coefficients of variation at the national, provincial/territorial, industrial, and industrial groups by province/territory levels. The sample was also inflated to compensate for dead, non-responding, and misclassified units.

MRTS is a repeated survey with maximisation of monthly sample overlap. The sample is kept month after month, and every month new units are added (births) to the sample.  MRTS births, i.e., new clusters of establishment(s), are identified every month via the BR’s latest universe. They are stratified according to the same criteria as the initial population. A sample of these births is selected according to the sampling fraction of the stratum to which they belong and is added to the monthly sample. Deaths occur on a monthly basis. A death can be a cluster of establishment(s) that have ceased their activities (out-of-business) or whose major activities are no longer in retail trade (out-of-scope). The status of these businesses is updated on the BR using administrative sources and survey feedback, including feedback from the MRTS. Methods to treat dead units and misclassified units are part of the sample and population update procedures.

5. Questionnaire design

The Monthly Retail Trade Survey incorporates the following sub-surveys:

Monthly Retail Trade Survey - R8

Monthly Retail Trade Survey (with inventories) – R8

Survey of Sales and Inventories of Alcoholic Beverages

The questionnaires collect monthly data on retail sales and the number of trading locations by province or territory and inventories of goods owned and intended for resale from a sample of retailers. The items on the questionnaires have remained unchanged for several years. For the 2004 redesign, the general questionnaires were subject to cosmetic changes only. The questionnaire for Sales and Inventories of Alcoholic Beverages underwent more extensive changes. The modifications were discussed with stakeholders and the respondents were given an opportunity to comment before the new questionnaire was finalized. If further changes are needed to any of the questionnaires, proposed changes would go through a review committee and a field test with respondents and data users to ensure its relevancy.

6. Response and nonresponse

6.1. Response and non-response

Despite the best efforts of survey managers and operations staff to maximize response in the MRTS, some non-response will occur. For statistical establishments to be classified as responding, the degree of partial response (where an accurate response is obtained for only some of the questions asked a respondent) must meet a minimum threshold level below which the response would be rejected and considered a unit nonresponse.  In such an instance, the business is classified as not having responded at all.

Non-response has two effects on data: first it introduces bias in estimates when nonrespondents differ from respondents in the characteristics measured; and second, it contributes to an increase in the sampling variance of estimates because the effective sample size is reduced from that originally sought.

The degree to which efforts are made to get a response from a non-respondent is based on budget and time constraints, its impact on the overall quality and the risk of nonresponse bias.

The main method to reduce the impact of non-response at sampling is to inflate the sample size through the use of over-sampling rates that have been determined from similar surveys.

Besides the methods to reduce the impact of non-response at sampling and collection, the non-responses to the survey that do occur are treated through imputation. In order to measure the amount of non-response that occurs each month, various response rates are calculated. For a given reference month, the estimation process is run at least twice (a preliminary and a revised run). Between each run, respondent data can be identified as unusable and imputed values can be corrected through respondent data. As a consequence, response rates are computed following each run of the estimation process.

For the MRTS, two types of rates are calculated (un-weighted and weighted). In order to assess the efficiency of the collection process, un-weighted response rates are calculated. Weighted rates, using the estimation weight and the value for the variable of interest, assess the quality of estimation. Within each of these types of rates, there are distinct rates for units that are surveyed and for units that are only modeled from administrative data that has been extracted from GST files.

To get a better picture of the success of the collection process, two un-weighted rates called the ‘collection results rate’ and the ‘extraction results rate’ are computed. They are computed by dividing the number of respondents by the number of units that we tried to contact or tried to receive extracted data for them. Non-monthly reporters (respondents with special reporting arrangements where they do not report every month but for whom actual data is available in subsequent revisions) are excluded from both the numerator and denominator for the months where no contact is performed.

In summary, the various response rates are calculated as follows:

Weighted rates:

Survey Response rate (estimation) =
Sum of weighted sales of units with response status i / Sum of survey weighted sales

where i = units that have either reported data that will be used in estimation or are converted refusals, or have reported data that has not yet been resolved for estimation.

Admin Response rate (estimation) =
Sum of weighted sales of units with response status ii / Sum of administrative weighted sales

where ii = units that have data that was extracted from administrative files and are usable for estimation.

Total Response rate (estimation) =
Sum of weighted sales of units with response status i or response status ii / Sum of all weighted sales

Un-weighted rates:

Survey Response rate (collection) =
Number of questionnaires with response status iii/ Number of questionnaires with response status iv

where iii = units that have either reported data (unresolved, used or not used for estimation) or are converted refusals.

where iv = all of the above plus units that have refused to respond, units that were not contacted and other types of non-respondent units.

Admin Response rate (extraction) =
Number of questionnaires with response status vi/ Number of questionnaires with response status vii

where vi = in-scope units that have data (either usable or non-usable) that was extracted from administrative files

where vii = all of the above plus units that have refused to report to the administrative data source, units that were not contacted and other types of non-respondent units.

(% of questionnaire collected over all in-scope questionnaires)

Collection Results Rate =
Number of questionnaires with response status iii / Number of questionnaires with response status viii

where iii = same as iii defined above

where viii = same as iv except for the exclusion of units that were contacted because their response is unavailable for a particular month since they are non-monthly reporters.

Extraction Results Rate =
Number of questionnaires with response status ix / Number of questionnaires with response status vii

where ix = same as vi with the addition of extracted units that have been imputed or were out of scope

where vii = same as vii defined above

(% of questionnaires collected over all questionnaire in-scope we tried to collect)

All the above weighted and un-weighted rates are provided at the industrial group, geography and size group level or for any combination of these levels.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden and survey costs, especially for smaller businesses, the MRTS has reduced the number of simple establishments in the sample that are surveyed directly and instead derives sales data for these establishments from Goods and Service Tax (GST) files using a statistical model. The model accounts for differences between sales and revenue (reported for GST purposes) as well as for the time lag between the survey reference period and the reference period of the GST file.

For more information on the methodology used for modeling sales from administrative data sources, refer to ‘Monthly Retail Trade Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

Table 1 contains the weighted response rates for all industry groups as well as for total retail trade for each province and territory. For more detailed weighted response rates, please contact the Marketing and Dissemination Section at (613) 951-3549, toll free: 1-877-421-3067 or by e-mail at retailinfo@statcan.

6.2. Methods used to reduce non-response at collection

Significant effort is spent trying to minimize non-response during collection. Methods used, among others, are interviewer techniques such as probing and persuasion, repeated re-scheduling and call-backs to obtain the information, and procedures dealing with how to handle non-compliant (refusal) respondents.

If data are unavailable at the time of collection, a respondent's best estimates are also accepted, and are subsequently revised once the actual data become available.

To minimize total non-response for all variables, partial responses are accepted. In addition, questionnaires are customized for the collection of certain variables, such as inventory, so that collection is timed for those months when the data are available.

Finally, to build trust and rapport between the interviewers and respondents, cases are generally assigned to the same interviewer each month. This action establishes a personal relationship between interviewer and respondent, and builds respondent trust.

7. Data collection and capture operations

Collection of the data is performed by Statistics Canada’s Regional Offices.

Table 1: Weighted response rates by NAICS, for all provinces/territories: November 2011

	Weighted Response Rates
	Total	Survey	Administrative
NAICS - Canada
Motor Vehicle and Parts Dealers	93.4	94.1	64.2
Automobile Dealers	95.7	96.2	52.6
New Car Dealers1	97.2	97.2
Used Car Dealers	73.1	77.5	52.6
Other Motor Vehicle Dealers	68.2	65.6	86.9
Automotive Parts, Accessories and Tire Stores	83.8	86.3	63
Furniture and Home Furnishings Stores	86.4	89.8	55.8
Furniture Stores	90	91.3	60.3
Home Furnishings Stores	80.7	86.9	54.2
Electronics and Appliance Stores	84.4	86.4	28.4
Building Material and Garden Equipment Dealers	86.7	90.3	58.1
Food and Beverage Stores	87	93.3	19.6
Grocery Stores	86.6	93.8	17.7
Grocery (except Convenience) Stores	89.5	96.6	18
Convenience Stores	51.6	58.1	15.5
Specialty Food Stores	68.9	77.3	33.2
Beer, Wine and Liquor Stores	93	94.8	22.9
Health and Personal Care Stores	89.5	91.1	71.7
Gasoline Stations	87.4	90	50.7
Clothing and Clothing Accessories Stores	86.1	87.7	40.8
Clothing Stores	86.2	87.5	44.9
Shoe Stores	92.1	93.2	34.9
Jewellery, Luggage and Leather Goods Stores	78.6	82.8	29.2
Sporting Goods, Hobby, Book and Music Stores	88.1	91.8	34.9
General Merchandise Stores	99	99.7	15.4
Department Stores	100	100
Other general merchadise stores	97.9	99.3	15.4
Miscellaneous Store Retailers	77.1	83.5	22.1
Total	90	92.8	41.6
Regions
Newfoundland and Labrador	87.4	88.7	35.7
Prince Edward Island	90.6	92.2
Nova Scotia	93.8	95.6	52.1
New Brunswick	88.7	91.1	57.9
Québec	89.1	93.7	34.3
Ontario	91.4	94.2	41.4
Manitoba	86.6	87.2	55.9
Saskatchewan	91	93.9	34.8
Alberta	88.2	90.1	52.9
British Columbia	89.8	92.3	43.3
Yukon Territory	86.7	86.7
Northwest Territories	85	85
Nunavut	90.9	90.9
1. There are no administrative records used in new car dealers

Weighted Response Rates

Respondents are sent a questionnaire or are contacted by telephone to obtain their sales and inventory values, as well as to confirm the opening or closing of business trading locations. Collection of the data begins approximately 7 working days after the end of the reference month and continues for the duration of that month.

New entrants to the survey are introduced to the survey via an introductory letter that informs the respondent that a representative of Statistics Canada will be calling. This call is to introduce the respondent to the survey, confirm the respondent's business activity, establish and begin data collection, as well as to answer any questions that the respondent may have.

8. Editing

Data editing is the application of checks to detect missing, invalid or inconsistent entries or to point to data records that are potentially in error. In the survey process for the MRTS, data editing is done at two different time periods.

First of all, editing is done during data collection. Once data are collected via the telephone, or via the receipt of completed mail-in questionnaires, the data are captured using customized data capture applications. All data are subjected to data editing. Edits during data collection are referred to as field edits and generally consist of validity and some simple consistency edits. They are used to detect mistakes made during the interview by the respondent or the interviewer and to identify missing information during collection in order to reduce the need for follow-up later on. Another purpose of the field edits is to clean up responses. In the MRTS, the current month’s responses are edited against the respondent’s previous month’s responses and/or the previous year’s responses for the current month. Field edits are also used to identify problems with data collection procedures and the design of the questionnaire, as well as the need for more interviewer training.

Follow-up with respondents occurs to validate potential erroneous data following any failed preliminary edit check of the data. Once validated, the collected data is regularly transmitted to the head office in Ottawa.

Secondly, editing known as statistical editing is also done after data collection and this is more empirical in nature. Statistical editing is run prior to imputation in order to identify the data that will be used as a basis to impute non-respondents. Large outliers that could disrupt a monthly trend are excluded from trend calculations by the statistical edits. It should be noted that adjustments are not made at this stage to correct the reported outliers.

The first step in the statistical editing is to identify which responses will be subjected to the statistical edit rules. Reported data for the current reference month will go through various edit checks.

The first set of edit checks is based on the Hidiriglou-Berthelot method whereby a ratio of the respondent’s current month data over historical (last month, same month last year) or auxiliary data is analyzed. When the respondent’s ratio differs significantly from ratios of respondents who are similar in terms of industry and/or geography group, the response is deemed an outlier.

The second set of edits consists of an edit known as the share of market edit. With this method, one is able to edit all respondents, even those where historical and auxiliary data is unavailable. The method relies on current month data only. Therefore, within a group of respondents, that are similar in terms of industrial group and/or geography, if the weighted contribution of a respondent to the group’s total is too large, it will be flagged as an outlier.

For edit checks based on the Hidiriglou-Berthelot method, data that are flagged as an outlier will not be included in the imputation models (those based on ratios). Also, data that are flagged as outliers in the share of market edit will not be included in the imputation models where means and medians are calculated to impute for responses that have no historical responses.

In conjunction with the statistical editing after data collection of reported data, there is also error detection done on the extracted GST data. Modeled data based on the GST are also subject to an extensive series of processing steps which thoroughly verify each record that is the basis for the model as well as the record being modeled. Edits are performed at a more aggregate level (industry by geography level) to detect records which deviate from the expected range, either by exhibiting large month-to-month change, or differing significantly from the remaining units. All data which fail these edits are subject to manual inspection and possible corrective action.

9. Imputation

Imputation in the MRTS is the process used to assign replacement values for missing data. This is done by assigning values when they are missing on the record being edited to ensure that estimates are of high quality and that a plausible, internal consistency is created. Due to concerns of response burden, cost and timeliness, it is generally impossible to do all follow-ups with the respondents in order to resolve missing responses. Since it is desirable to produce a complete and consistent microdata file, imputation is used to handle the remaining missing cases.

In the MRTS, imputation is based on historical data or administrative data (GST sales). The appropriate method is selected according to a strategy that is based on whether historical data is available, auxiliary data is available and/or which reference month is being processed.

There are three types of historical imputation methods. The first type is a general trend that uses one historical data source (previous month, data from next month or data from same month previous year). The second type is a regression model where data from previous month and same month previous year are used simultaneously. The third type uses the historical data as a direct replacement value for a non-respondent. Depending upon the particular reference month, there is an order of preference that exists so that top quality imputation can result. The historical imputation method that was labelled as the third type above is always the last option in the order for each reference month.

The imputation methods using administrative data are automatically selected when historical information is unavailable for a non-respondent. The administrative data source (annual GST sales) is the basis of these methods. The annual GST sales are used for two types of methods. One is a general trend that will be used for simple structure, e.g. enterprises with only one establishment, and a second type is called median-average that is used for units with a more complex structure.

10. Estimation

Estimation is a process that approximates unknown population parameters using only part of the population that is included in a sample. Inferences about these unknown parameters are then made, using the sample data and associated survey design. This stage uses Statistics Canada's Generalized Estimation System (GES).

For retail sales, the population is divided into a survey portion (take-all and take-some strata) and a non-survey portion (take-none stratum). From the sample that is drawn from the survey portion, an estimate for the population is determined through the use of a Horvitz-Thompson estimator where responses for sales are weighted by using the inverses of the inclusion probabilities of the sampled units. Such weights (called sampling weights) can be interpreted as the number of times that each sampled unit should be replicated to represent the entire population. The calculated weighted sales values are summed by domain, to produce the total sales estimates by each industrial group / geographic area combination. A domain is defined as the most recent classification values available from the BR for the unit and the survey reference period. These domains may differ from the original sampling strata because units may have changed size, industry or location. Changes in classification are reflected immediately in the estimates and do not accumulate over time. For the non-survey portion, the sales are estimated with statistical models using monthly GST sales.

For more information on the methodology for modeling sales from administrative data sources which also contributes to the estimates of the survey portion, refer to ‘Monthly Retail Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

The measure of precision used for the MRTS to evaluate the quality of a population parameter estimate and to obtain valid inferences is the variance. The variance from the survey portion is derived directly from a stratified simple random sample without replacement.

Sample estimates may differ from the expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

11. Revisions and seasonal adjustment

Revisions in the raw data are required to correct known non-sampling errors. These normally include replacing imputed data with reported data, corrections to previously reported data, and estimates for new births that were not known at the time of the original estimates. Raw data are revised, on a monthly basis, for the month immediately prior to the current reference month being published. That is, when data for December are being published for the first time, there will also be revisions, if necessary, to the raw data for November. In addition, revisions are made once a year, with the initial release of the February data, for all months in the previous year. The purpose is to correct any significant problems that have been found that apply for an extended period. The actual period of revision depends on the nature of the problem identified, but rarely exceeds three years. Time series contain the elements essential to the description, explanation and forecasting of the behaviour of an economic phenomenon: "They are statistical records of the evolution of economic processes through time."¹ Economic time series such as the Monthly Retail Trade Survey can be broken down into five main components: the trend-cycle, seasonality, the trading-day effect, the Easter holiday effect and the irregular component.

The trend represents the long-term change in the series, whereas the cycle represents a smooth, quasi-periodical movement about the trend, showing a succession of growth and decline phases (e.g., the business cycle). These two components—the trend and the cycle—are estimated together, and the trend-cycle reflects the fundamental evolution of the series. The other components reflect short-term transient movements.

The seasonal component represents sub-annual, monthly or quarterly fluctuations that recur more or less regularly from one year to the next. Seasonal variations are caused by the direct and indirect effects of the climatic seasons and institutional factors (attributable to social conventions or administrative rules; e.g., Christmas).

The trading-day component originates from the fact that the relative importance of the days varies systematically within the week and that the number of each day of the week in a given month varies from year to year. This effect is present when activity varies with the day of the week. For instance, Sunday is typically less active than the other days, and the number of Sundays, Mondays, etc., in a given month changes from year to year.

The Easter holiday effect is the variation due to the shift of part of April’s activity to March when Easter falls in March rather than April.

Lastly, the irregular component includes all other more or less erratic fluctuations not taken into account in the preceding components. It is a residual that includes errors of measurement on the 1. A Note on the Seasonal adjustment of Economic Time Series», Canadian Statistical Review, August 1974.  A variable itself as well as unusual events (e.g., strikes, drought, floods, major power blackout or other unexpected events causing variations in respondents’ activities).

Thus, the latter four components—seasonal, irregular, trading-day and Easter holiday effect—all conceal the fundamental trend-cycle component of the series. Seasonal adjustment (correction of seasonal variation) consists in removing the seasonal, trading-day and Easter holiday effect components from the series, and it thus helps reveal the trend-cycle. While seasonal adjustment permits a better understanding of the underlying trend-cycle of a series, the seasonally adjusted series still contains an irregular component. Slight month-to-month variations in the seasonally adjusted series may be simple irregular movements. To get a better idea of the underlying trend, users should examine several months of the seasonally adjusted series.

Since April 2008, Monthly Retail Trade Survey data are seasonally adjusted using the X-12- ARIMA2 software. The technique that is used essentially consists of first correcting the initial series for all sorts of undesirable effects, such as the trading-day and the Easter holiday effects, by a module called regARIMA. These effects are estimated using regression models with ARIMA errors (auto-regressive integrated moving average models). The series can also be extrapolated for at least one year by using the model. Subsequently, the raw series—pre-adjusted and extrapolated if applicable— is seasonally adjusted by the X-11 method.

The X-11 method is used for analysing monthly and quarterly series. It is based on an iterative principle applied in estimating the different components, with estimation being done at each stage using adequate moving averages3. The moving averages used to estimate the main components—the trend and seasonality—are primarily smoothing tools designed to eliminate an undesirable component from the series. Since moving averages react poorly to the presence of atypical values, the X-11 method includes a tool for detecting and correcting atypical points. This tool is used to clean up the series during the seasonal adjustment. Outlying data points can also be detected and corrected in advance, within the regARIMA module.

Lastly, the annual totals of the seasonally adjusted series are forced to the annual totals of the original series.

Unfortunately, seasonal adjustment removes the sub-annual additivity of a system of series; small discrepancies can be observed between the sum of seasonally adjusted series and the direct seasonal adjustment of their total. To insure or restore additivity in a system of series, a reconciliation process is applied or indirect seasonal adjustment is used, i.e. the seasonal adjustment of a total is derived by the summation of the individually seasonally adjusted series.

12. Data quality evaluation

The methodology of this survey has been designed to control errors and to reduce their potential effects on estimates. However, the survey results remain subject to errors, of which sampling error is only one component of the total survey error. Sampling error results when observations are made only on a sample and not on the entire population. All other errors arising from the various phases of a survey are referred to as nonsampling errors. For example, these types of errors can occur when a respondent provides incorrect information or does not answer certain questions; when a unit in the target population is omitted or covered more than once; when GST data for records being modeled for a particular month are not representative of the actual record for various reasons; when a unit that is out of scope for the survey is included by mistake or when errors occur in data processing, such as coding or capture errors.

Prior to publication, combined survey results are analyzed for comparability; in general, this includes a detailed review of individual responses (especially for large businesses), general economic conditions and historical trends.

A common measure of data quality for surveys is the coefficient of variation (CV). The coefficient of variation, defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. Since the coefficient of variation is calculated from responses of individual units, it also measures some non-sampling errors.

The formula used to calculate coefficients of variation (CV) as percentages is:

CV (X) = S(X) * 100% / X
where X denotes the estimate and S(X) denotes the standard error of X.

Confidence intervals can be constructed around the estimates using the estimate and the CV. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a CV of 2%, the standard error will be $240,000 (the estimate multiplied by the CV). It can be stated with 68% confidence that the expected values will fall within the interval whose length equals the standard deviation about the estimate, i.e. between $11,760,000 and $12,240,000.

Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e. between $11,520,000 and $12,480,000.

Finally, due to the small contribution of the non-survey portion to the total estimates, bias in the non-survey portion has a negligible impact on the CVs. Therefore, the CV from the survey portion is used for the total estimate that is the summation of estimates from the surveyed and non-surveyed portions.

13. 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.

Confidentiality analysis includes the detection of possible "direct disclosure", which occurs when the value in a tabulation cell is composed of a few respondents or when the cell is dominated by a few companies.

The Monthly Survey of Manufacturing (MSM) publishes statistical series for manufacturers – sales of goods manufactured, inventories, unfilled orders and new orders. The values of these characteristics represent current monthly estimates of the more complete Annual Survey of Manufactures and Logging (ASML) data.

The MSM is a sample survey of approximately 10,500 Canadian manufacturing establishments, which are categorized into over 220 industries. Industries are classified according to the 2007 North American Industrial Classification System (NAICS). Seasonally adjusted series are available for the main aggregates.

An establishment comprises the smallest manufacturing unit capable of reporting the variables of interest. Data collected by the MSM provides a current ‘snapshot’ of sales of goods manufactured values by the Canadian manufacturing sector, enabling analysis of the state of the Canadian economy, as well as the health of specific industries in the short- to medium-term. The information is used by both private and public sectors including Statistics Canada, federal and provincial governments, business and trade entities, international and domestic non-governmental organizations, consultants, the business press and private citizens. The data are used for analyzing market share, trends, corporate benchmarking, policy analysis, program development, tax policy and trade policy.

1. Sales of goods manufactured

Sales of goods manufactured (formerly shipments of goods manufactured) are defined as the value of goods manufactured by establishments that have been shipped to a customer. Sales of goods manufactured exclude any wholesaling activity, and any revenues from the rental of equipment or the sale of electricity. Note that in practice, some respondents report financial trans­ac­tions rather than payments for work done. Sales of goods manufactured are available by 3-digit NAICS, for Canada and broken down by province.

For the aerospace product and parts, and shipbuilding industries, the value of production is used instead of sales of goods manufactured. This value is calculated by adjusting monthly sales of goods manufactured by the monthly change in inventories of goods / work in process and finished goods manufactured. Inventories of raw materials and components are not included in the calculation since production tries to measure "work done" during the month. This is done in order to reduce distortions caused by the sales of goods manufactured of high value items as completed sales.

2. Inventories

Measurement of component values of inventory is important for economic studies as well as for derivation of production values. Respondents are asked to report their book values (at cost) of raw materials and components, any goods / work in process, and fin­ished goods manufactured inventories separately. In some cases, respondents estimate a total inventory figure, which is allocated on the basis of proportions reported on the ASML. Inventory levels are calculated on a Canada‑wide basis, not by province.

3. Orders

a) Unfilled Orders

Unfilled orders represent a backlog or stock of orders that will generate future sales of goods manufactured assuming that they are not cancelled. As with inventories, unfilled orders and new orders levels are calculated on a Canada‑wide basis, not by province.

The MSM produces estimates for unfilled orders for all industries except for those industries where orders are customarily filled from stocks on hand and order books are not gen­erally maintained. In the case of the aircraft companies, options to purchase are not treated as orders until they are entered into the account­ing system.

b) New Orders

New orders represent current demand for manufactured products. Estimates of new orders are derived from sales of goods manufactured and unfilled orders data. All sales of goods manufactured within a month result from either an order received during the month or at some earlier time. New orders can be calculated as the sum of sales of goods manufactured adjusted for the monthly change in unfilled orders.

4. Non-Durable / Durable goods

a) Non-durable goods industries include:

Food (NAICS 311),
Beverage and Tobacco Products (312),
Textile Mills (313),
Textile Product Mills (314),
Clothing (315),
Leather and Allied Products (316),
Paper (322),
Printing and Related Support Activities (323),
Petroleum and Coal Products (324),
Chemicals (325) and
Plastic and Rubber Products (326).

b) Durable goods industries include:

Wood Products (NAICS 321),
Non-Metallic Mineral Products (327),
Primary Metals (331),
Fabricated Metal Products (332),
Machinery (333),
Computer and Electronic Products (334),
Electrical Equipment, Appliance and Components (335),
Transportation Equipment (336),
Furniture and Related Products (337) and
Miscellaneous Manufacturing (339). 

Survey design and methodology

Beginning with the August 1999 reference month, the Monthly Survey of Manufacturing (MSM) underwent an extensive redesign.

Concept Review

In 1998, it was decided that before any redesign work could begin the basic concepts and definitions of the program would be confirmed.

This was done in two ways: First, a review of user requirements was initiated. This involved revisiting an internal report to ensure that the user requirements from that exercise were being satisfied. As well, another round of internal review with the major users in the National Accounts was undertaken. This was to specifically focus on any data gaps that could be identified.

Secondly, with these gaps or requirements in hand, a survey was conducted in order to ascertain respondent’s ability to report existing and new data. The study was also to confirm that respondents understood the definitions, which were being asked by survey analysts.

The result of the concept review was a reduction of the number of questions for the survey from sixteen to seven. Most of the questions that were dropped had to do with the reporting of sales of goods manufactured for work that was partially completed.

In 2007, the MSM terminology was updated to be Charter of Accounts (COA) compliant. With the August 2007 reference month release the MSM has harmonized its concepts to the ASML. The variable formerly called “Shipments” is now called “Sales of goods manufactured”. As well, minor modifications were made to the inventory component names. The definitions have not been modified nor has the information collected from the survey.

Methodology

The latest sample design incorporates the 2007 North American Industrial Classification Standard (NAICS). Stratification is done by province with equal quality requirements for each province. Large size units are selected with certainty and small units are selected with a probability based on the desired quality of the estimate within a cell.

The estimation system generates estimates using the NAICS. The estimates will also continue to be reconciled to the ASML. Provincial estimates for all variables will be produced. A measure of quality (CV) will also be produced.

Components of the Survey Design

Target Population and Sampling Frame

Statistics Canada’s business register provides the sampling frame for the MSM. The target population for the MSM consists of all statistical establishments on the business register that are classified to the manufacturing sector (by NAICS). The sampling frame for the MSM is determined from the target population after subtracting establishments that represent the bottom 5% of the total manufacturing sales of goods manufactured estimate for each province. These establishments were excluded from the frame so that the sample size could be reduced without significantly affecting quality.

The Sample

The MSM sample is a probability sample comprised of approximately 10,500 establishments. A new sample was chosen in the autumn of 2006, followed by a six-month parallel run (from reference month September 2006 to reference month February 2007). The refreshed sample officially became the new sample of the MSM effective in January 2007.

This marks the first process of refreshing the MSM sample since 2002. The objective of the process is to keep the sample frame as fresh and up-to date as possible. All establishments in the sample are refreshed to take into account changes in their value of sales of goods manufactured, the removal of dead units from the sample and some small units are rotated out of the GST-based portion of the sample, while others are rotated into the sample.

Prior to selection, the sampling frame is subdivided into industry-province cells. For the most part, NAICS codes were used. Depending upon the number of establishments within each cell, further subdivisions were made to group similar sized establishments’ together (called stratum). An establishment’s size was based on its most recently available annual sales of goods manufactured or sales value. 

Each industry by province cell has a ‘take-all’ stratum composed of establishments sampled each month with certainty. This ‘take-all’ stratum is composed of establishments that are the largest statistical enterprises, and have the largest impact on estimates within a particular industry by province cell. These large statistical enterprises comprise 45% of the national manufacturing sales of goods manufactured estimates.

Each industry by province cell can have at most three ‘take-some’ strata. Not all establishments within these stratums need to be sampled with certainty. A random sample is drawn from the remaining strata. The responses from these sampled establishments are weighted according to the inverse of their probability of selection. In cells with take-some portion, a minimum sample of 10 was imposed to increase stability.

The take-none portion of the sample is now estimated from administrative data and as a result, 100% of the sample universe is covered. Estimation of the take-none portion also improved efficiency as a larger take-none portion was delineated and the sample could be used more efficiently on the smaller sampled portion of the frame.

Data Collection

Only a subset of the sample establishments is sent out for data collection. For the remaining units, information from administrative data files is used as a source for deriving sales of goods manufactured data. For those establishments that are surveyed, data collection, data capture, preliminary edit and follow-up of non-respondents are all performed in Statistics Canada regional offices. Sampled establishments are contacted by mail or telephone according to the preference of the respondent. Data capture and preliminary editing are performed simultaneously to ensure the validity of the data.

In some cases, combined reports are received from enterprises or companies with more than one establishment in the sample where respondents prefer not to provide individual establishment reports. Businesses, which do not report or whose reports contain errors, are followed up immediately.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden, especially for small businesses, Statistics Canada has been investigating various alternatives to survey taking. Administrative data files are a rich source of information for business data and Statistics Canada is working at mining this rich data source to its full potential. As such, effective the August 2004 reference month, the MSM reduced the number of simple establishments in the sample that are surveyed directly and instead, derives sales of goods manufactured data for these establishments from Goods and Services Tax (GST) files using a statistical model. The model accounts for the difference between sales of goods manufactured (reported to MSM) and sales (reported for GST purposes) as well as the time lag between the reference period of the survey and the reference period of the GST file.

In conjunction with the most recent sample, effective January 2007, approximately 2,500 simple establishments were selected to represent the GST portion of the sample.

Inventories and unfilled orders estimates for establishments where sales of goods manufactured are GST-based are derived using the MSM’s imputation system. The imputation system applies to the previous month values, the month-to-month and year-to-year changes in similar firms which are surveyed. With the most recent sample, the eligibility rules for GST-based establishments were refined to have more GST-based establishments in industries that typically carry fewer inventories. This way the impact of the GST-based establishments which require the estimation of inventories, will be kept to a minimum.

Detailed information on the methodology used for modelling sales of goods manufactured from administrative data sources can be found in the ‘Monthly Survey of Manufacturing: Use of Administrative Data’ (Catalogue no. 31-533-XIE) document.

Data quality

Statistical Edit and Imputation

Data are analyzed within each industry-province cell. Extreme values are listed for inspection by the magnitude of the deviation from average behavior. Respondents are contacted to verify extreme values. Records that fail statistical edits are considered outliers and are not used for imputation.

Values are imputed for the non-responses, for establishments that do not report or only partially complete the survey form. A number of imputation methods are used depending on the variable requiring treatment. Methods include using industry-province cell trends, historical responses, or reference to the ASML. Following imputation, the MSM staff performs a final verification of the responses that have been imputed.

Revisions

In conjunction with preliminary estimates for the current month, estimates for the previous three months are revised to account for any late returns. Data are revised when late responses are received or if an incorrect response was recorded earlier.

Estimation

Estimates are produced based on returns from a sample of manufacturing establishments in combination with administrative data for a portion of the smallest establishments. The survey sample includes 100% coverage of the large manufacturing establishments in each industry by province, plus partial coverage of the medium and small-sized firms. Combined reports from multi-unit companies are pro-rated among their establishments and adjustments for progress billings reflect revenues received for work done on large item contracts. Approximately 2,500 of the sampled medium and small-sized establishments are not sent questionnaires, but instead their sales of goods manufactured are derived by using revenue from the GST files. The portion not represented through sampling – the take-none portion - consist of establishments below specified thresholds in each province and industry. Sub-totals for this portion are also derived based on their revenues.

Industry values of sales of goods manufactured, inventories and unfilled orders are estimated by first weighting the survey responses, the values derived from the GST files and the imputations by the number of establishments each represents. The weighted estimates are then summed with the take-none portion. While sales of goods manufactured estimates are produced by province, no geographical detail is compiled for inventories and orders since many firms cannot report book values of these items monthly.

Benchmarking

Up to and including 2003, the MSM was benchmarked to the Annual Survey of Manufactures and Logging (ASML). Benchmarking was the regular review of the MSM estimates in the context of the annual data provided by the ASML. Benchmarking re-aligned the annualized level of the MSM based on the latest verified annual data provided by the ASML.

Significant research by Statistics Canada in 2006 to 2007 was completed on whether the benchmark process should be maintained. The conclusion was that benchmarking of the MSM estimates to the ASML should be discontinued. With the refreshing of the MSM sample in 2007, it was determined that benchmarking would no longer be required (retroactive to 2004) because the MSM now accurately represented 100% of the sample universe. Data confrontation will continue between MSM and ASML to resolve potential discrepancies. 

As of the January 2007 reference month, a new sample was introduced. It is standard practice that every few years the sample is refreshed to ensure that the survey frame is up to date with births, deaths and other changes in the population. The refreshed sample is linked at the detailed level to prevent data breaks and to ensure the continuity of time series. It is designed to be more representative of the manufacturing industry at both the national and provincial levels.

Data confrontation and reconciliation

Each year, during the period when the Annual Survey of Manufactures and Logging section set their annual estimates, the MSM section works with the ASML section to confront and reconcile significant differences in values between the fiscal ASML and the annual MSM at the strata and industry level.

The purpose of this exercise of data reconciliation is to highlight and resolve significant differences between the two surveys and to assist in minimizing the differences in the micro-data between the MSM and the ASML.

Sampling and Non-sampling Errors

The statistics in this publication are estimates derived from a sample survey and, as such, can be subject to errors. The following material is provided to assist the reader in the interpretation of the estimates published.

Estimates derived from a sample survey are subject to a number of different kinds of errors. These errors can be broken down into two major types: sampling and non-sampling.

1. Sampling Errors

Sampling errors are an inherent risk of sample surveys. They result from the difference between the value of a variable if it is randomly sampled and its value if a census is taken (or the average of all possible random values). These errors are present because observations are made only on a sample and not on the entire population.

The sampling error depends on factors such as the size of the sample, variability in the population, sampling design and method of estimation. For example, for a given sample size, the sampling error will depend on the stratification procedure employed, allocation of the sample, choice of the sampling units and method of selection. (Further, even for the same sampling design, we can make different calculations to arrive at the most efficient estimation procedure.) The most important feature of probability sampling is that the sampling error can be measured from the sample itself.

2. Non-sampling Errors

Non-sampling errors result from a systematic flaw in the structure of the data-collection procedure or design of any or all variables examined. They create a difference between the value of a variable obtained by sampling or census methods and the variable’s true value. These errors are present whether a sample or a complete census of the population is taken. Non-sampling errors can be attributed to one or more of the following sources:

a) Coverage error: This error can result from incomplete listing and inadequate coverage of the population of interest.

b) Data response error: This error may be due to questionnaire design, the characteristics of a question, inability or unwillingness of the respondent to provide correct information, misinterpretation of the questions or definitional problems.

c) Non-response error: Some respondents may refuse to answer questions, some may be unable to respond, and others may be too late in responding. Data for the non-responding units can be imputed using the data from responding units or some earlier data on the non-responding units if available.

The extent of error due to imputation is usually unknown and is very much dependent on any characteristic differences between the respondent group and the non-respondent group in the survey. This error generally decreases with increases in the response rate and attempts are therefore made to obtain as high a response rate as possible.

d) Processing error: These errors may occur at various stages of processing such as coding, data entry, verification, editing, weighting, and tabulation, etc. Non-sampling errors are difficult to measure. More important, non-sampling errors require control at the level at which their presence does not impair the use and interpretation of the results.

Measures have been undertaken to minimize the non-sampling errors. For example, units have been defined in a most precise manner and the most up-to-date listings have been used. Questionnaires have been carefully designed to minimize different interpretations. As well, detailed acceptance testing has been carried out for the different stages of editing and processing and every possible effort has been made to reduce the non-response rate as well as the response burden.

Measures of Sampling and Non-sampling Errors

1. Sampling Error Measures

The sample used in this survey is one of a large number of all possible samples of the same size that could have been selected using the same sample design under the same general conditions. If it was possible that each one of these samples could be surveyed under essentially the same conditions, with an estimate calculated from each sample, it would be expected that the sample estimates would differ from each other.

The average estimate derived from all these possible sample estimates is termed the expected value. The expected value can also be expressed as the value that would be obtained if a census enumeration were taken under identical conditions of collection and processing. An estimate calculated from a sample survey is said to be precise if it is near the expected value.

Sample estimates may differ from this expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

The standard error is a measure of precision in absolute terms. The coefficient of variation (CV), defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. For comparison purposes, one may more readily compare the sampling error of one estimate to the sampling error of another estimate by using the coefficient of variation.

In this publication, the coefficient of variation is used to measure the sampling error of the estimates. However, since the coefficient of variation published for this survey is calculated from the responses of individual units, it also measures some non-sampling error.

The formula used to calculate the published coefficients of variation (CV) in Table 1 is:

CV(X) = S(X)/X

where X denotes the estimate and S(X) denotes the standard error of X.

In this publication, the coefficient of variation is expressed as a percentage.

Confidence intervals can be constructed around the estimate using the estimate and the coefficient of variation. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a coefficient of variation of 10%, the standard error will be $1,200,000 or the estimate multiplied by the coefficient of variation. It can then be stated with 68% confidence that the expected value will fall within the interval whose length equals the standard deviation about the estimate, i.e., between $10,800,000 and $13,200,000. Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e., between $9,600,000 and $14,400,000.

Text table 1 contains the national level CVs, expressed as a percentage, for all manufacturing for the MSM characteristics. For CVs at other aggregate levels, contact the Dissemination and Frame Services Section at (613) 951-9497, toll free: 1-866-873-8789 or by e-mail at manufact@statcan.gc.ca.

Text table 1
National Level CVs by Characteristic

Month	Sales of goods manufactured %	Raw materials and components inventories %	Goods / work in process inventories %	Finished goods manufactured inventories %	Unfilled Orders %
Nov-10	0.84	1.16	1.62	1.44	1.72
Dec-10	0.75	1.19	1.62	1.42	1.70
Jan-11	0.80	1.20	1.68	1.35	1.68
Feb-11	0.74	1.22	1.72	1.38	1.93
Mar-11	0.74	1.21	1.66	1.33	2.77
Apr-11	0.76	1.20	1.73	1.33	2.70
May-11	0.77	1.20	1.71	1.40	2.67
Jun-11	0.77	1.16	1.76	1.41	2.73
Jul-11	0.74	1.19	1.80	1.41	2.64
Aug-11	0.78	1.26	1.87	1.38	2.62
Sep-11	0.80	1.30	1.87	1.38	2.61
Oct-11	0.83	1.31	1.93	1.35	2.66
Nov-11	0.85	1.32	1.81	1.35	2.97

2. Non-sampling Error Measures

The exact population value is aimed at or desired by both a sample survey as well as a census. We say the estimate is accurate if it is near this value. Although this value is desired, we cannot assume that the exact value of every unit in the population or sample can be obtained and processed without error. Any difference between the expected value and the exact population value is termed the bias. Systematic biases in the data cannot be measured by the probability measures of sampling error as previously described. The accuracy of a survey estimate is determined by the joint effect of sampling and non-sampling errors.

Three sources of non-sampling error in the MSM are non-response error, imputation error and the error due to editing. To assist users in evaluating these errors, weighted rates that are related to these three types of error are given in Text table 2. The following is an example of what is meant by a weighted rate. A cell with a sample of 20 units in which five respond for a particular month would have a response rate of 25%. If these five reporting units represented $8 million out of a total estimate of $10 million, the weighted response rate would be 80%.

The definitions of the three weighted rates noted in Text table 2 follow. The weighted response rate is the proportion of a characteristic’s total estimate that is based upon reported data (excluding data that has been edited). The weighted imputation rate is the proportion of a characteristic’s total estimate that is based upon imputed data. The weighted editing rate is the proportion of a characteristic’s total estimate that is based upon data that was edited (edited data may have been originally reported or imputed).

Text table 2 contains the three types of weighted rates for each of the characteristics at the national level for all of manufacturing. In the table, the rates are expressed as percentages.

Text Table 2
National Weighted Rates by Source and Characteristic

Characteristics	Survey Source			Administrative Data Source
	Response	Imputation	Editing	Modeled	Imputation	Editing
	%	%	%	%	%	%
Sales of goods manufactured	83.64	3.99	5.56	6.18	0.41	0.22
Raw materials and components	73.63	8.25	9.16	0.00	8.94	0.00
Goods / work in process	59.36	10.48	23.08	0.00	6.56	0.51
Finished goods manufactured	77.86	7.87	4.73	0.00	8.83	0.72
Unfilled Orders	49.35	6.30	39.76	0.00	3.33	1.26

Joint Interpretation of Measures of Error

The measure of non-response error as well as the coefficient of variation must be considered jointly to have an overview of the quality of the estimates. The lower the coefficient of variation and the higher the weighted response rate, the better will be the published estimate.

Seasonal Adjustment

Economic time series contain the elements essential to the description, explanation and forecasting of the behavior of an economic phenomenon. They are statistical records of the evolution of economic processes through time. In using time series to observe economic activity, economists and statisticians have identified four characteristic behavioral components: the long-term movement or trend, the cycle, the seasonal variations and the irregular fluctuations. These movements are caused by various economic, climatic or institutional factors. The seasonal variations occur periodically on a more or less regular basis over the course of a year. These variations occur as a result of seasonal changes in weather, statutory holidays and other events that occur at fairly regular intervals and thus have a significant impact on the rate of economic activity.

In the interest of accurately interpreting the fundamental evolution of an economic phenomenon and producing forecasts of superior quality, Statistics Canada uses the X12-ARIMA seasonal adjustment method to seasonally adjust its time series. This method minimizes the impact of seasonal variations on the series and essentially consists of adding one year of estimated raw data to the end of the original series before it is seasonally adjusted per se. The estimated data are derived from forecasts using ARIMA (Auto Regressive Integrated Moving Average) models of the Box-Jenkins type.

The X-12 program uses primarily a ratio-to-moving average method. It is used to smooth the modified series and obtain a preliminary estimate of the trend-cycle. It also calculates the ratios of the original series (fitted) to the estimates of the trend-cycle and estimates the seasonal factors from these ratios. The final seasonal factors are produced only after these operations have been repeated several times.

The technique that is used essentially consists of first correcting the initial series for all sorts of undesirable effects, such as the trading-day and the Easter holiday effects, by a module called regARIMA. These effects are then estimated using regression models with ARIMA errors. The series can also be extrapolated for at least one year by using the model. Subsequently, the raw series, pre-adjusted and extrapolated if applicable, is seasonally adjusted by the X-12 method.

The procedures to determine the seasonal factors necessary to calculate the final seasonally adjusted data are executed every month. This approach ensures that the estimated seasonal factors are derived from an unadjusted series that includes all the available information about the series, i.e. the current month's unadjusted data as well as the previous month's revised unadjusted data.

While seasonal adjustment permits a better understanding of the underlying trend-cycle of a series, the seasonally adjusted series still contains an irregular component. Slight month-to-month variations in the seasonally adjusted series may be simple irregular movements. To get a better idea of the underlying trend, users should examine several months of the seasonally adjusted series.

The aggregated Canada level series are now seasonally adjusted directly, meaning that the seasonally adjusted totals are obtained via X-12-ARIMA. Afterwards, these totals are used to reconcile the provincial total series which have been seasonally adjusted individually.

For other aggregated series, indirect seasonal adjustments are used. In other words, their seasonally adjusted totals are derived indirectly by the summation of the individually seasonally adjusted kinds of business.

Trend

A seasonally adjusted series may contain the effects of irregular influences and special circumstances and these can mask the trend. The short term trend shows the underlying direction in seasonally adjusted series by averaging across months, thus smoothing out the effects of irregular influences. The result is a more stable series. The trend for the last month may be, subject to significant revision as values in future months are included in the averaging process.

Real manufacturing sales of goods manufactured, inventories, and orders

Changes in the values of the data reported by the Monthly Survey of Manufacturing (MSM) may be attributable to changes in their prices or to the quantities measured, or both. To study the activity of the manufacturing sector, it is often desirable to separate out the variations due to price changes from those of the quantities produced. This adjustment is known as deflation.

Deflation consists in dividing the values at current prices obtained from the survey by suitable price indexes in order to obtain estimates evaluated at the prices of a previous period, currently the year 2002. The resulting deflated values are said to be “at 2002 prices”. Note that the expression “at current prices” refer to the time the activity took place, not to the present time, nor to the time of compilation.

The deflated MSM estimates reflect the prices that prevailed in 2002. This is called the base year. The year 2002 was chosen as base year since it corresponds to that of the price indexes used in the deflation of the MSM estimates. Using the prices of a base year to measure current activity provides a representative measurement of the current volume of activity with respect to that base year. Current movements in the volume are appropriately reflected in the constant price measures only if the current relative importance of the industries is not very different from that in the base year.

The deflation of the MSM estimates is performed at a very fine industry detail, equivalent to the 6-digit industry classes of the North American Industry Classification System (NAICS). For each industry at this level of detail, the price indexes used are composite indexes which describe the price movements for the various groups of goods produced by that industry.

With very few exceptions the price indexes are weighted averages of the Industrial Product Price Indexes (IPPI). The weights are derived from the annual Canadian Input-Output tables and change from year to year. Since the Input-Output tables only become available with a delay of about two and a half years, the weights used for the most current years are based on the last available Input-Output tables.

The same price index is used to deflate sales of goods manufactured, new orders and unfilled orders of an industry. The weights used in the compilation of this price index are derived from the output tables, evaluated at producer’s prices. Producer prices reflect the prices of the goods at the gate of the manufacturing establishment and exclude such items as transportation charges, taxes on products, etc. The resulting price index for each industry thus reflects the output of the establishments in that industry.

The price indexes used for deflating the goods / work in process and the finished goods manufactured inventories of an industry are moving averages of the price index used for sales of goods manufactured. For goods / work in process inventories, the number of terms in the moving average corresponds to the duration of the production process. The duration is calculated as the average over the previous 48 months of the ratio of end of month goods / work in process inventories to the output of the industry, which is equal to sales of goods manufactured plus the changes in both goods / work in process and finished goods manufactured inventories.

For finished goods manufactured inventories, the number of terms in the moving average reflects the length of time a finished product remains in stock. This number, known as the inventory turnover period, is calculated as the average over the previous 48 months of the ratio of end-of-month finished goods manufactured inventory to sales of goods manufactured.

To deflate raw materials and components inventories, price indexes for raw materials consumption are obtained as weighted averages of the IPPIs. The weights used are derived from the input tables evaluated at purchaser’s prices, i.e. these prices include such elements as wholesaling margins, transportation charges, and taxes on products, etc. The resulting price index thus reflects the cost structure in raw materials and components for each industry.

The raw materials and components inventories are then deflated using a moving average of the price index for raw materials consumption. The number of terms in the moving average corresponds to the rate of consumption of raw materials. This rate is calculated as the average over the previous four years of the ratio of end-of-year raw materials and components inventories to the intermediate inputs of the industry.