The informative value of central banks talks: a topic model application to sentiment analysis
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
<abstract><p>Central banks communication has lately become an important tool to guide expectations and its impact on the economy has been acknowledged by the literature. Nowadays central banks speeches face an increasing variety of topics, which are not discriminated by text analysis. In this paper we build a topic-weighted central bank sentiment index as a combination of machine learning and text analysis techniques to investigate large datasets. First, we develop a methodological framework to grid search the best Latent Dirichlet Allocation (LDA) model to uncover the latent topics in central banks' speeches and releases published between 2000 and 2021. Then, we build a topic-specific sentiment index based on dictionary techniques. Next, we summarise the results in a topic-weighted Central Bank Sentiment Index (CBSIw) for the Bank of Canada (BoC), the Bank of England (BoE), the European Central Bank (ECB) and the Federal Reserve (Fed). We find that the main common driver of the CBSIw is the monetary policy topic, followed by macroprudential policy and payments and settlements. We also uncover bank-specific topics and topics related to new challenges, for example innovation and climate change. Moreover, we find that the CBSIw decreases after the Great Recession, signalling a worsening in sentiment, as well as during the COVID-19 crisis. Finally, we employ a probit regression to further assess the predictive power of our monetary policy topic-specific index. We find that the indicator helps predicting future changes in policy rate, corroborating the evidence that central banks communication signals future monetary policy decisions.</p></abstract>
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.
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 | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it