What indicators are projected? The selection of indicators varies by source, but often includes:
- GDP and its components
- CPI or other inflation indicators
- Unemployment rate and other labor market indicators
Many of the reports from Eurosystem member banks include technical assumptions of the projection, such as:
- Exchange rates
- Interest rates
- Commodity prices
- Migration
These reports rarely contain raw levels. Rather, the indicators have been transformed to an analytical form, such as the following. To use these projections effectively, we strongly recommend that you consult the reports published by each source; you may need to examine both a chart and data table.
- Annual percent change
- Percent share of GDP
- Percent contribution to growth in GDP
Are there variants of each indicator? If a source publishes several scenarios, we carry only the baseline.
If the source publishes a range instead of a single value, we carry the upper and lower values in separate series; if both a range and a central value, we carry the latter.
Some sources project the performance not only of their own country, but of major trading partners and blocs. The Bank of Canada, for example, projects GDP for the U.S., euro area, oil-importing emerging-market economies, and others. We carry these additional countries.
Are the projections subject to revision? Yes. Past and future periods are subject to revision, depending on the source. We intend to update the Data Buffet extracts periodically, but not more often than once per month. (The Central Bank of Brazil updates its projections daily, but we will harvest values as of the last day of each month.)
We will not announce such changes on Data Buffet News with a "data revision" article. Instead, you can check the advance release calendar in Mnemonic 411, subscribe to notifications through Mnemonic 411 (see related article), or create a basket schedule with an "any change to series" trigger (see related article).
So that you can track the evolution of the projections, we will store each version in vintaged time series (see below). The most recent version will be stored under the canonical mnemonic; hence, a saved basket will operate without interruption.
What are the mnemonics like? The generic structure is:
J (^^ theme) (^ measure) (^^ source) (^ indicator) (^ adjustment) (^ frequency) . (I^^^ economy)
The source is the country of the central bank (expressed as the ISO 3166 alpha-2 country code), and the economy is the Data Buffet standard geo code.
For example:
- J NA % GR GDP U A.IGRC = Growth rate of GDP for Greece, by Greece
- J NA % CA GDP U A.IUSA = Growth rate of GDP for U.S., by Canada
- J NA % CA INF U A.ICAN = CPI inflation for Canada, by Canada
- J LW Q DE UN W A.IDEU = Unemployed persons (count, working day adjusted) for Germany, by Germany
- J PC N IE OIL U A.IIRL = Oil price (in USD), by Ireland
To facilitate cross-national retrieval, we have standardized the mnemonics so far as practical, in part by omitting certain definitional nuances ("distinctions without a difference"); hence, similar mnemonics do not guarantee comparable measurements. Instead, these nuances are expressed in the unit-descriptor metadata (see below).
These concept codes generally adhere to the symbology found elsewhere in Data Buffet, but not universally. To select series, please use the catalog.
How do I retrieve related series? Use wild card expressions such as:
- J^^^CA?.ICAN = all projections by Canada for Canada
- J^^^CA?.? = all projections by Canada for all regions
- J?.ICAN = all projections for Canada
- JNA^^^GDP?.? = all GDP projections by all sources
How do I retrieve the vintages? We will store successive prior versions of each series under mnemonics that contain an additional vintage specifier, consisting of the four-digit year and two-digit month that the projection was published (and prefixed with an underscore). For example, if it appeared in a quarterly bulletin published in April 2017, the segment would be _201704. The above mnemonic structure is thus extended:
J (theme) (meas) (source) (indicator) _yyyymm (adj) (freq) . (economy)
For example:
- JNA%CAGDPUA.IUSA = Growth rate of GDP for U.S., by Canada
- JNA%CAGDP_201609UA.IUSA = 2016m09 vintage
- JNA%CAGDP_201704UA.IUSA = 2017m04 vintage
- JNA%CAGDP_2017^^UA.IUSA = All vintages for 2017 (wild card expression)
Each vintage will be stored in a distinct databank, which you will need to activate (see related article). For example, the current vintage of the above projection resides in ICAN_BOC_FOR.db, and the hypothetical 2016m09 vintage in ICAN_BOC_FOR_201609.db. The databank is listed in Mnemonic 411.
What kinds of measurements are there, i.e., what do the unit-descriptors look like? Most of these reports are concerned with growth of GDP and other measures, rather than the underlying level; this is usually expressed as an annual percent change. Because the analytical form varies by source, please consult the unit-descriptor. Moreover, there will be more variety in the technical assumptions. For example:
- (% Y/Y) = year-over-year percent change, as with annual growth
- (% pt.) = percentage points, as with contribution to growth
- (% AVE) = percent, average over period
- (USD per EUR) = exchange rate, U.S. dollar per euro
For specifics, please consult the individual "new data" articles.
Where are the projections stored? In each country catalog, under (Cyclical indicators » Outlook).
Which countries and sources? At present, the project includes these national sources:
Updates
- 2 Jun 2017 - Initial version
- 25 Jan 2019 - Add: Greece, U.S. CBO
- 23 May 2019 - Add: Brazil, India, Japan
- 25 Feb 2020 - Add: Euro zone, South Africa, USDA