Statistical Confidence Intervals for the Bank of Canada's Business Outlook Survey
Bibliographic record
Abstract
While a number of central banks publish their own business conditions indicators that rely on non-random sampling, knowledge about their statistical accuracy has been limited. Recently, de Munnik, Dupuis, and Illing (2009) made some progress in this area for the Bank of Canada's Business Outlook Survey (BOS) by estimating the impact of the Bank's non-random sampling on the accuracy of the survey results. They found no evidence that the Bank's firm-selection process results in significantly biased estimates and/or wider confidence intervals than in the random-selection case. The author deepens and extends this work by (i) outlining the statistical properties of population-proportion and balance-of-opinion questions, and demonstrating how their design affects the calculation of the confidence intervals; (ii) examining the variation in statistical confidence associated with changes in the underlying response distribution using actual quarterly BOS results; (iii) considering the possibility that statistical accuracy varies across questions; and (iv) investigating whether the statistical accuracy of the survey results changes with variations in the business cycle. The main findings are that confidence intervals around the population-proportion questions are about half of those for the balance-of-opinion questions, and that the confidence bands around both types of question can change from survey to survey when the underlying response distribution becomes more or less concentrated in particular response categories (such as “higher,” “the same,” or “lower”). The author finds that confidence intervals around the BOS population-proportion questions become somewhat narrower during periods of recession, while those for the balance-of-opinion questions vary within a similar range across the cycle.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".