Claim and evidence Close ×
ECB Working Paper No. 352 of April 2004, by Paul McNelis and Peter McAdam, reports that “thick” models combining several neural-network models outperform the best-performing linear models in real-time and bootstrap inflation forecasts for euro-area service-price indices, and do as well or sometimes better for consumer and producer price indices across several countries. Copy claim link
Copy claim reference
Machine-checked Stated in source Supported
Supporting evidence: primary · 1 source; Context: primary · 1 source
Date 2004-04 · publication Records dated 2004 → Role Earliest located Register (kind of text) Nonfiction Scope and limits A staff working paper under the ECB disclaimer; one co-author is outside the ECB. It uses neural networks, a machine-learning technique, without the words “machine learning” or “artificial intelligence”, and reports results for particular indices and periods, not general superiority (the non-technical summary says linear and neural-network forecasts converge in periods without structural change).
The evidence 03 primary source Supports the claim
Forecasting inflation with thick models and neural networks ↗ European Central Bank · 2004-04
Copy source reference
Cite the original source Cite the original document, not Selekoa. Check these details against it.
Copy APA
Download BibTeX Download RIS Download CSL-JSON
Source excerpt “Thick models represent “trimmed mean” forecasts from several neural network models. They outperform the best performing linear models for “real-time” and “bootstrap” forecasts for service indices for the euro area, and do well, sometimes better, for the more general consumer and producer price indices across a variety of countries.” Location in source PDF p. 5, abstract
How it was read Retrieved directly from the publisher
Machine-checked · 2026-11-06
Source details and check notes Authors Paul McNelis; Peter McAdam Version Retrieved for the candidate pack on 2026-11-04 (SHA-256 4252bd524ee1f3ccc3a2a90541ff30556c5f75ed77c2244c78e56cfdb0e32dce). Retrieved again at admission on 2026-11-06 (same file). PDF of ECB Working Paper No. 352 (April 2004) as served by the ECB Canonical address Open the canonical source ↗ Accessed 2026-11-06 Language English (en) Attribution Paul McNelis, Peter McAdam, “Forecasting inflation with thick models and neural networks”, 2004-04. Reuse No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution. Check notes Candidate pack passage, AI and money (research/ai-money). The quotation (49 words) matched a fresh retrieval of 2026-11-06 (the same file the candidate pack read). Matching ignored differences of whitespace, case, quotation marks, dashes, line-break hyphens and Unicode compatibility forms. No human editorial review. primary source Context only
Forecasting inflation with thick models and neural networks ↗ European Central Bank · 2004-04
Copy source reference
Cite the original source Cite the original document, not Selekoa. Check these details against it.
Copy APA
Download BibTeX Download RIS Download CSL-JSON
Source excerpt “We show in this paper that the linear and neural network specifications converge in their forecasts in such periods.” Location in source PDF p. 6, non-technical summary
How it was read Retrieved directly from the publisher
Machine-checked · 2026-11-06
Source details and check notes Authors Paul McNelis; Peter McAdam Version Retrieved for the candidate pack on 2026-11-04 (SHA-256 4252bd524ee1f3ccc3a2a90541ff30556c5f75ed77c2244c78e56cfdb0e32dce). Retrieved again at admission on 2026-11-06 (same file). PDF of ECB Working Paper No. 352 (April 2004) as served by the ECB Canonical address Open the canonical source ↗ Accessed 2026-11-06 Language English (en) Attribution Paul McNelis, Peter McAdam, “Forecasting inflation with thick models and neural networks”, 2004-04. Reuse No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution. Check notes Candidate pack passage, AI and money (research/ai-money). The quotation (19 words) matched a fresh retrieval of 2026-11-06 (the same file the candidate pack read). Matching ignored differences of whitespace, case, quotation marks, dashes, line-break hyphens and Unicode compatibility forms. No human editorial review. primary source Context only
Forecasting inflation with thick models and neural networks ↗ European Central Bank · 2004-04
Copy source reference
Cite the original source Cite the original document, not Selekoa. Check these details against it.
Copy APA
Download BibTeX Download RIS Download CSL-JSON
Source excerpt “The views expressed in this paper do not necessarily reflect those of the European Central Bank.” Location in source PDF p. 3, imprint
How it was read Retrieved directly from the publisher
Machine-checked · 2026-11-06
Source details and check notes Authors Paul McNelis; Peter McAdam Version Retrieved for the candidate pack on 2026-11-04 (SHA-256 4252bd524ee1f3ccc3a2a90541ff30556c5f75ed77c2244c78e56cfdb0e32dce). Retrieved again at admission on 2026-11-06 (same file). PDF of ECB Working Paper No. 352 (April 2004) as served by the ECB Canonical address Open the canonical source ↗ Accessed 2026-11-06 Language English (en) Attribution Paul McNelis, Peter McAdam, “Forecasting inflation with thick models and neural networks”, 2004-04. Reuse No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution. Check notes Candidate pack passage, AI and money (research/ai-money). The text’s own statement of whose views it gives, shown beside the record; it is context, not support. The quotation matched a fresh retrieval of 2026-11-06. No human editorial review. Status and meaning Editorial status. Machine-checked . Machines found this record and checked it against its sources; no person accepted it individually. Each evidence record above shows how it was checked.
Relation. A passage of the document meets the proposition’s criterion. The record documents what the text states, in its register (fiction and satire are labelled); it is not evidence that the proposition is true and implies no influence on or from another text.
Date. Date of publication, as the text gives it.
Record ID am-u05-ecb-wp-352