Yayue Deng1,Mohan Xu2,Yao Tang2,Yao Tang is the corresponding author.
Abstract
The effectiveness of central bank communication is a crucial aspect of monetary policy transmission. While recent research has examined the influence of policy communication by the chairs of the Federal Reserve on various financial variables, much of the literature relies on rule-based or dictionary-based methods in parsing the language of the chairs, leaving nuanced information about policy stance contained in nonverbal emotion out of the analysis. In the current study, we propose the Fine-Grained Monetary Policy Analysis Framework (FMPAF), a novel approach that integrates large language models (LLMs) with regression analysis to provide a comprehensive analysis of the impact of the press-conference communications of chairs of the Federal Reserve on financial markets. We conduct extensive comparisons of model performance under different levels of granularity, modalities, and communication scenarios. Based on our preferred specification, a one-unit increase in the sentiment score is associated with an increase of the price of S&P 500 Exchange-Traded Fund by approximately 500 basis points, a 15-basis-point decrease in the policy interest rate, while not leading to a significant response in exchange rates.
Introduction
Volatility in the financial market is driven by numerous factors, such as macroeconomic policies, public confidence, and news releases. The complex relationship among these variables renders the forecast of financial variables exceedingly challenging, as effective forecasting depends heavily on the gathering and parsing of a vast array of external information. Among various sources, information released by central banks is particularly significant, as their decisions and guidance have a profound and widespread impact on the economy and the financial market (Tsukioka and Yamasaki 2020; Bennani etal. 2020; Shah, Paturi, and Chava 2023). For instance, in August 2020, Jerome H. Powell, the Chairman of the Federal Reserve (the Fed), delivered a high-profile speech at the Jackson Hole Symposium, where he announced a major shift in the Fed’s approach toward inflation. Powell indicated that the Fed would be willing to permit inflation to exceed the conventional 2% target for a period of time before hiking interest rates. This announcement led to immediate and notable reactions in the stock market. The S&P 500 and Nasdaq indices surged to a record high on impact, as investors interpreted the Fed’s current stance as a signal for prolonged low-interest rates, which typically benefit the stock market.
Because central banks are well known to exert heavy influence on the financial market, several AI-driven works in the finance literature (Gorodnichenko, Pham, and Talavera 2023; Mathur etal. 2022; Shah, Paturi, and Chava 2023; Matsui, Ren, and Ferrara 2021; Zirn, Meusel, and Stuckenschmidt 2015; Frunza 2020) concentrate on central banks’ policy communication with the public, especially communication at the Fed’s Federal Open Market Committee (FOMC) meeting. A recent work (Mathur etal. 2022) constructs a multimodal dataset, MONOPOLY, from press conference videos for the purpose of financial forecasting. Matsui, Ren, and Ferrara (2021) propose a word embedding-based framework for understanding the policy communication of the Fed. Zirn, Meusel, and Stuckenschmidt (2015) use graph clustering methods to unsupervisedly reveal opinion groups in transcripts of the FOMC discussions. Frunza (2020) introduce an unsupervised approach to extract economic concept-value pairs from FOMC statements. However, these methods either utilize traditional dictionary-based sentiment analysis (Tsukioka and Yamasaki 2020; Narain and Sangani 2023), or only focus on the pre-processing or construction of data sources (Shah, Paturi, and Chava 2023; Frunza 2020; Mathur etal. 2022), without analyzing the effects of information on the financial market. The proposed methods are often not validated with empirical tests.
In terms of the first limitation of sentiment analysis, the challenge in understanding policy stance lies in quantifying its subtle cues, which are often soft and implicit information. Previous studies mainly employ dictionary-based or rule-based methods, attempting to quantify textual data through pre-defined emotion lexicons or rules, respectively. Dictionary-based sentiment analysis suffers from a limited vocabulary and inaccurate performance compared to large language models (LLMs). Besides, the complexity of policy stances extends far beyond the dimensions of emotion, necessitating more nuanced analysis methods for a comprehensive understanding. Secondly, the analysis of financial market responses is under-explored in these studies (Zirn, Meusel, and Stuckenschmidt 2015; Frunza 2020; Mathur etal. 2022; Shah, Paturi, and Chava 2023), as only classification accuracy is reported. In addition, Rozkrut etal. (2007) points out that the effects of central banks’ communication on the financial market vary with the sources of communication, such as press conference post FOMC meetings and speech by Fed officials at other venues. Therefore, it is valuable to investigate the effect of different sources of communication on macroeconomic variables.
To address the limitations discussed above, we introduce a Fine-grained Monetary Policy Analysis Framework (FMPAF) that adopts a powerful pre-trained language model (PLMs) to enhance comprehension capability and fine-tune it in the monetary policy stance corpus to transfer learning ability to the specific policy domain. Leveraging the FMPAF, our research delves into the financial market responses to Fed communication from different sources, including our personally collected semi-annual monetary policy hearings and press conferences. Moreover, we employ a policy indicator to elucidate the impulse response of stock prices, exchange rates, and interest rates in the financial markets of the US and other important economies.
The current study is closely related to Gorodnichenko, Pham, and Talavera (2023), which also focuses on the policy communication of the Fed in the form of textual and acoustic materials, but their study relies on analysis at the conversation level. Such a coarse-grained analysis can lead to the loss of information. In contrast, we employ fine-grained sentiment analysis and conduct multi-modal analysis at the sentence level. Our comparative analysis reveals that our fine-grained method enhances predictability, revealing a significant correlation with financial variables.
In summary, the proposed FMPAF aims to establish a correlation between shifts in policy sentiment and subsequent market movements by quantifying sentiment in the communications made by chairs of the Fed. Our framework applies PLMs-based models from both the textual and acoustic sides to detect the fine-grained policy stance embedded in the language of the chairs. Moreover, our empirical tests suggest a high degree of validity for the FMPAF, as the framework generates a reliable indicator of the chairs’ underlying policy stance. Extensive comprehensive analysis demonstrates that the indicator of the policy stance is predictive of the fluctuation of the financial market.
Background and Related Work
Sentiment Analysis in Finance
The term Sentiment Analysis was first defined by Nasukawa et al. (Nasukawa and Yi 2003) as determining the subjectivity polarity (positive or negative) and polarity strength (strongly positive, mildly positive, weakly positive, etc.) of a given review text. Sentiment analysis is increasingly recognized as a pivotal task within the financial domain, closely linked with market sentiment, which subsequently influences price movements. Early literature (Bharathi and Geetha 2017; Gorodnichenko, Pham, and Talavera 2023) predominantly focused on word categorization methods, such as the bag of words technique, to assess tonality. The introduction of the polarity lexicon by (Loughran and Mcdonald 2010) stands out as a significant advancement in the development of sentiment analysis methodologies for finance. Further, (Garcia, Hu, and Rohrer 2023) contributes new lexicons of positive and negative terms specifically for financial contexts. Nonetheless, these dictionary-based approaches face two principal challenges: 1) Context Insensitivity: They often neglect the subtleties of context, such as sarcasm or culturally specific expressions, resulting in imprecise sentiment analysis; 2) Inadequacy in Processing Complex Emotions: These methods are limited in their ability to interpret complex emotional expressions and fail to accurately discern sentiment intensities or nuances.Recently, large language models like Bert (Alaparthi and Mishra 2020) and RoBERTa (Liu etal. 2019) have demonstrated extensive capabilities, significantly enhancing a range of natural language processing (NLP) tasks. Despite their remarkable efficacy in a wide array of NLP applications, Pretrained Language Models (PLMs) encounter limitations in domain-specific tasks. This is primarily due to the marked disparity between the distribution of general training data and the specialized language of specific domains (Gururangan etal. 2020; Gu etal. 2022). An expression that holds a strong semantic orientation in one domain might necessitate a different interpretation in another. To mitigate this, several financial language models (Araci 2019; Yang, Uy, and Huang 2020) have been proposed to enhance performance in analyzing financial and economic texts. For example, recent endeavors, such as by (Shah, Paturi, and Chava 2023), have concentrated on monetary orientation analysis through the fine-tuning of large language models.
Policy Communication of Central Bank
Economists have long recognized the importance of central bank communication (Kohn and Sack 2003), making efforts to study its impact on financial markets both quantitatively and qualitatively (Blinder etal. 2008; Blinder 2018; Neuhierl and Weber 2019; Coenen etal. 2017). Despite all these efforts, the communication mechanism between the Fed Chairs and the market, an important channel of central bank communication, has not been fully explored and understood due to the technical limitations. Most related papers are restricted to the impact of the Chairs’ communication on the domestic financial market (Gorodnichenko, Pham, and Talavera 2023; Hansen and McMahon 2016). To fill in the gap, our paper investigates the impact of the Chairs’ language on the international financial market for the very first time, which contributes to the broad literature on cross-border transmission of monetary policy.
Source Years File Modalities File Avg. Words Avg. Duration (s) Total (words & s) Panel A: Raw Data Policy Reports 2003-2023 Readout Textual 42 2376.1 - 92708 Speeches Acoustic 30 - 9018.6 261540.4 Transcribed text Textual 30 21059.5 - 610726.0 Press Conferences 2011-2022 Speeches Acoustic 47 - 3379.3 160845.6 Transcribed text Textual 47 8445.4 - 405379.0 Panel B: Processed Data Policy Reports 2003-2023 Readout Textual 3230 28.7 - 92708 Speeches Acoustic 36203 - 6.1 219963.3 Transcribed text Textual 36210 16.8 - 610726.0 Press Conferences 2011-2022 Speeches Acoustic 19025 - 8.2 158826.5 Transcribed text Textual 19440 20.4 - 405379.0
Methodology
As shown in Fig1, the proposed FMPAF contains four parts: 1) Dataset Collection, 2) Preprocessing, 3) Deep-learning-based multi-modal Analysing, and 4) Economic impact analysis via indicators. In this section, we first describe the background of our data resource. Besides, we illustrate our deep-learning-based method to conduct acoustic and linguistic analysis for investigating the economic impact of policy communication.
Problem formulation
In order to investigate the impact of the Chairman’s communication on financial markets, we formulate the problem as a regression task. Following (Jordà 2005), the impulse response of financial market and the Fed’s communication can be formulated as the local projection specification as follows:
(1) |
where denotes the date of the press conferences, measures the policy stance embedded in the opening remarks and Q&A sessions, , , and capture the policy action of the Fed.
Controls | Sources |
---|---|
FFR Shock | Swanson (2021) |
FG Shock | |
AP Shock | |
Shadow Rate | Wu and Xia (2016) |
To avoid the omitted variable problem, we introduce four control variables into our regression, as shown in Table2. Therefore, the impact of on financial outcomes can be clearly identified.For the financial outcomes, we employ stock prices, exchange rates, and interest rates, using the log difference between close and open prices to measure the accumulated return, which is referred to as the . Take the representative stock price in the USA (SPY) as an example, we compute the according to:
(2) |
We estimate the above specification for each horizon in a daily frequency, ranging from 0 day to 15 days. To illustrate the dynamic effects explicitly, we draw the estimated coefficients. 90 % bias-corrected confidence intervals are reported, shown by the dash lines in each figure.
Dataset Collection and Preprocessing
Our dataset comprises two primary types of data:
- 1.
The Fed’s monetary policy hearings: presented to Congress by the Chair on monetary policy and economic outlook, including both a written report and an oral testimony at a Congressional hearing, conducted formally and systematically.
- 2.
The Fed’s press conferences: held after FOMC meetings, where the Chair announces monetary policy decisions and answers journalists’ questions.
Both types of data are essential for how the Fed communicates its monetary policy to the public and legislative bodies.
Monetary Policy HearingsOur dataset of this part includes samples from the United States Senate Committee on Banking, Housing, and Urban Affairs website111https://www.banking.senate.gov/hearings/, detailing hearings on financial and economic policies. An example is the July 13, 2017222https://www.banking.senate.gov/hearings/the-semiannual-monetary-policy-report-to-the-congress, hearing on the semiannual monetary policy report, featuring Janet L. Yellen, then Chair of the Fed, as the key witness. The website provides dates, times, locations, and additional resources like opening remarks and testimonies.
In our study, we scraped the United States Senate Committee on Banking, Housing, and Urban Affairs website to compile a list of all hearings. We then applied two criteria to identify the Monetary Policy Hearings: 1) the title must contain “monetary policy report,” and 2) the “Chair” must be listed among the attendees. From this, we extracted essential details from the hearing URLs, such as primary attendees, audio recordings, the chair’s testimonies, and meeting dates. For the formal statement documents (readouts) reported to Congress, we used Pdfplumber (Singer-Vine and The pdfplumber contributors 2023), a Python package, to extract the text and segment it into sentences. For meetings without transcripts, we employed an Automatic Speech Recognition (ASR) tool, WhisperX (Bain etal. 2023), for transcription. Our final dataset, as summarized in Table1, includes 42 Monetary Policy Meetings from 2003 to 2023, each with a PDF of the readout and, when available, an MP3 audio file. Notably, audio recordings were not available for all meetings; 12 out of 42 lacked audio files. We segmented a total of 36,203 audio pieces, averaging about 6.1 seconds each, mostly ranging from 3 to 6 seconds, showing a uniform distribution in duration.
Press ConferenceThe Federal Open Market Committee (FOMC), the Fed’s key monetary policymaking body, consists of the Board of Governors’ seven members and the presidents of the twelve Reserve Banks. They meet approximately eight times a year to discuss economic conditions and develop strategies for maximum employment and price stability. Since 1994, and regularly after each meeting since May 1999, the FOMC issues statements outlining federal funds rate targets. These statements gained importance after December 2008 when the federal funds rate hit its effective lower bound. To improve transparency, Chairman Ben Bernanke began holding press conferences after four annual meetings starting in April 2011, a practice that expanded to after every meeting since 2019.
Our dataset, sourced from public records, includes conference call videos, audio recordings, and text transcripts from six international banks from 2011 to 2022. We specifically analyze Fed press conference videos from April 27, 2011, to January 26, 2022, covering 47 conferences. Differing from the conversation-level analysis in (Gorodnichenko, Pham, and Talavera 2023), we segment our text and audio data into sentences, rather than larger paragraphs, to preserve detailed information. This approach, focusing on sentence-level granularity, provides more precise semantic insights and variation, enhancing our model’s understanding and predictive accuracy. Our comparative analysis between sentence-level and conversation-level approaches shows that our fine-grained method is more effective. Besides, it also offers more context than phrases-based or words-based analysis. Our analysis reveals a higher proportion of neutral segments in the data, reflecting the cautious communication style of Fed Chairs, who often use neutral language to avoid unintended policy signals. This tendency towards neutrality is evident in our predictive results, aligning with the Chairs’ careful choice of words.
Acoustic and linguistic Analysis
Linguistic AnalysisPrevious studies have demonstrated that fine-tuned pre-trained language models (PLMs) significantly outperform both rule-based models and zero-shot models. Therefore, we employ the RoBERTa-large pre-trained language model as our foundational model, which we further fine-tuned for downstream classification tasks. Numerous studies have successfully fine-tuned PLMs for tasks related to emotion classification. However, our research diverges by aiming to extract policy-related information from text. Drawing inspiration from prior literature (Shah, Paturi, and Chava 2023), we do not employ traditional emotional characteristics such as neutral, positive, and negative. Instead, we fine-tuned the RoBERTa model using a recent open-source dataset (Shah, Paturi, and Chava 2023). This dataset comprises three categories: Dovish, Hawkish, and Neutral. Dovish sentences indicate an impending easing of monetary policy, whereas Hawkish sentences signal an upcoming tightening of monetary policy. Finally, The model achieved an average testing accuracy of 72.8% and an average testing F1 score of 72.4%, using a learning rate of 0.000001.
Acoustic AnalysisIn response to the absence of a speech database tailored to policy stances, we adopt the wav2vec2.0 model (Baevski etal. 2020) which is composed of multi-layer convolutional layers. In order to extract emotion-related acoustic features, the model is fine-tuned to predict the speaker’s emotional state on a subset of the Interactive Emotional Dyadic Motion Capture (IEMOCAP) (Busso etal. 2008) database, a prominent repository for English emotional speech, containing nine emotions. This database includes both scripted and improvisational sessions between two actors, making it a benchmark for English SER systems. It contains 12 hours of audiovisual recordings by ten actors, totaling 10,039 utterances, each annotated by three evaluators. The subset focuses on primary emotional states like neutral, happiness, sadness, and anger. The model achieved an average testing accuracy of 75.3% across these four emotion categories.
Policy Indicator
After feeding the audio into our model, each segment received a label of predicted sentiment (dovish, neutral or hawkish). We aggregate the labels from sentence level to conference level in a concise way to construct an indicator of the Chair’s policy stance during a certain press conference:
(3) |
The closer this value is to 1, the more dovish in indicates; conversely, the closer this value is to -1, the more hawkish in indicates.We aggregate the audio segments in a parallel manner:
(4) |
The closer this value is to 1, the more positive it indicates; conversely, the closer this value is to -1, the more negative it indicates.
Regression and Impact Analysis
Comparative Analysis of Different Granularity
To evaluate the efficacy of the fine-grained method we employ in our framework, We first visualize the distribution of the sentiment indicators’ results derived from both fine-grained and coarse-grained methods. The coarse-grained method means we calculate the value of the sentiment indicator based on the conversational level as in (Gorodnichenko, Pham, and Talavera 2023). As shown in Fig.3, our findings reveal a high correlation in trends of sentiment across different granularities, suggesting that indicators from varied granularities yield consistent policy implications. Notably, the indicator based on fine-grained data demonstrates smoother trends with fewer extremes compared to its coarse-grained counterpart. This observation aligns with the expectation that as public figures, Chairs tend to maintain prudence and consistency in their communications.Furthermore, we perform a local-projection regression analysis on U.S. stock prices (SPY) to assess the performance of different granularity levels in predicting financial market volatility. Our results, illustrated in Fig.2, reveal that incorporating a fine-grained sentiment indicator significantly enhances the predictive capability of our framework. The sentiment’s coefficient became notably significant at a 10% confidence level with a 10-day lag. In summary, employing a fine-grained analysis method improves the accuracy and predictability of our framework, contributing to the resolution of critical economic issues.
Analysis of Domain-Specific Fine-tuning
In this section, we explore the impact of domain-specific fine-tuning. We use RoBERTa (Liu etal. 2019) as a baseline, which has been fine-tuned for three-category emotion recognition (positive, negative, and neutral) (Barbieri etal. 2020), in contrast to policy stance. We adjust our regression equation to incorporate specific sentiment measures. Our findings show that the financial market’s responses vary significantly between Fig.1(a) and Fig.4. Notably, the sentiment index based on positive and negative classifications proved consistently insignificant, even with control variables, suggesting it is not a reliable predictor of market volatility. However, one unit increase in the sentiment index using dovish-hawkish classifications increases SPY by approximately 500 bp after a 10-day lag, underscoring the value of domain-specific tuning. Consequently, we argue that the lack of a policy-specific speech corpus hindered our ability to achieve the desired analytical results from audio data, as discussed in Sec.Comparative Analysis of Different Modalities.
Comparative Analysis of Different Communication Scenarios
In this study, we focus on the communication from two scenarios: the Q&A sessions at the post-FOMC press conferences and the semiannual monetary policy hearings. Both of these two channels play a vital role in the Fed’s policy communication: 1) the press conferences serve to prepare the public by shaping their expectations for the upcoming policy decisions (Ehrmann and Fratzscher (2007); Boguth, Grégoire, and Martineau (2019)), 2) the semiannual monetary policy hearings constitute an important communication channel between the Fed and Congress, by providing opportunities for the Fed to convey its policy intentions and economic outlook((Labonte and Makinen 2008)). The dynamics of sentiment orientation are presented in Fig.5. We observe that over a given period of time, the inclinations and dynamics of the monetary policy reflected in what the Chair said in both scenarios of meetings remain highly consistent. The analysis results align with the actual economic situation. For example, since 2010, the Fed’s monetary policy can be broadly categorized into the following stages: 1) From 2010 to 2014, the Fed supported economic recovery after the Great Financial Crisis with very low interest rates and quantitative easing, which could be reflected with the dovish sentiment result; 2) During 2015 and 2019, the sentiment tended to be a little bit more hawkish as the Fed tapered its asset purchases and gradually increased the federal funds rate; 3) Between 2020 and 2021, in response to the COVID-19 pandemic, the Fed cut rates to near zero and resumed quantitative easing to support the economy, as reflected in Fig.5 as a jump; 4) After 2021, as inflation pressures mounted, the Fed started discussing tapering asset purchases again, setting the stage for potential rate hikes, thus we can see a giant drop in all the four lines in Fig.5 accordingly, reflecting that the Chair Powell is getting much more hawkish in every talks he gave. In conclusion, our recognition of the policy stance is basically in line with the literature Bernanke (2020).
Meanwhile, some systematic differences are also shown in Fig.5. The sentiment of policy communication tends to be more neutral at press conferences, as the score of red line is closer to zero. We explain the reason that the Federal Reserve Act requires the Fed chair to report to Congress twice a year on monetary policy, provide explanations, and answer questions from members of Congress. The Chair needs to explain the Fed’s policy path in a transparent way so that he can assure Congress that he has been highly responsible, thus the sentiment value for other lines, besides the red one, can be volatile more according to the varying economic fundamentals. On the other hand, when communicating with the public, the Chair would try to be neutral to avoid unnecessary panic in the market. Thus, it’s natural that Chairs would sound relatively neutral in the press conferences.
Comparative Analysis of Different Modalities
To compare the effectiveness of our approach under different modalities, we conduct two parallel empirical regressions on text and audio modalities respectively, as shown in Fig.6. The result reveals that the Voice Tone index, calculated from acoustic sentiment tags, lacked statistical significance in forecasting the financial market. We speculate that the scarcity of speech corpus tailored to the financial sector notably hampers the method’s efficacy. Consequently, the model’s capacity to discern the underlying sentiment in the Fed Chair’s speech is compromised, thereby impeding its predictive power to financial market dynamics. Therefore, it is of significant importance to establish a speech dataset related to the financial sector in the future.
Analysis for Cross-border Transmission
The cross-border transmission of monetary policy is quite an important empirical issue in finance and economics. The existing literature in this field is mainly based on diverse interest-rate indicators to measure monetary policy ((Barbosa etal. 2018)), while the soft information channel represented by the Chair’s speeches has been underexplored. This paper considers the impact of the policy stance revealed in the speeches of the Fed Chair on other major economies, including China and the EU. We investigate the variance of three key financial variables, namely the stock price, the monetary-policy-related interest rate and the exchange rate between the national currency and the US dollars.
The impulse response results in Fig.7 show that the Fed Chairs’ language do have a spillover impact on the financial markets abroad. The sign of the coefficients is the same, while the magnitude diversify across China and the European Union. For China, the sentiment has an immediate impact on China’s short-term money market rates, leading to a 15 bp decrease within 10 days; the response of the stock market is slightly delayed, with over 500 bp increase in the stock price in about a week later; however, the sentiment does not show a significant impact on the exchange rate between US dollars and China RMB, which may because after the rapid and flexible adjustment of interest rates, the interest rate parity condition is met again, at which point the exchange rate does not need further adjustment. The responses of financial markets in the EU are basically consistent with China, whether in sign or magnitude. However, the duration of the spillover effect is different. In terms of the responses of the stock prices and interest rates, the effect diminishes in 10 days in China, while it continues to move the corresponding markets in EU in each horizon. In conclusion, the effect of the Chairs’ language would spill over to the financial markets in other countries, especially those major economies which are closely linked to the USA in such a globalized world.
Conclusion
In this paper, we propose a novel, fine-grained monetary policy framework, FMPAF, for analyzing Fed communications from various data sources. We utilize a pre-trained large language model, fine-tuning it with a monetary policy-oriented corpus to transfer knowledge to this specific policy domain. Besides, we employ a pre-trained acoustic model to identify paralinguistic information in the Fed’s communications. Our analysis has revealed that domain-specific fine-tuning plays a crucial role in achieving significant results. After applying domain-specific fine-tuning, our method also demonstrates marginal improvements in significance. Furthermore, we verify the spillover effect of the Fed Chairs’ language on the international financial market.
References
- Alaparthi and Mishra (2020)Alaparthi, S.; and Mishra, M. 2020.Bidirectional Encoder Representations from Transformers (BERT): A sentiment analysis odyssey.CoRR, abs/2007.01127.
- Araci (2019)Araci, D. 2019.FinBERT: Financial Sentiment Analysis with Pre-trained Language Models.CoRR, abs/1908.10063.
- Baevski etal. (2020)Baevski, A.; Zhou, Y.; Mohamed, A.; and Auli, M. 2020.wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations.In NeurIPS.
- Bain etal. (2023)Bain, M.; Huh, J.; Han, T.; and Zisserman, A. 2023.WhisperX: Time-Accurate Speech Transcription of Long-Form Audio.INTERSPEECH 2023.
- Barbieri etal. (2020)Barbieri, F.; Camacho-Collados, J.; EspinosaAnke, L.; and Neves, L. 2020.TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification.In Findings of the Association for Computational Linguistics: EMNLP 2020, 1644–1650. Online: Association for Computational Linguistics.
- Barbosa etal. (2018)Barbosa, L.; Bonfim, D.; Costa, S.; and Everett, M. 2018.Cross-border spillovers of monetary policy: What changes during a financial crisis?Journal of International Money and Finance, 89: 154–174.
- Bennani etal. (2020)Bennani, H.; Fanta, N.; Gertler, P.; and Horvath, R. 2020.Does central bank communication signal future monetary policy in a (post)-crisis era? The case of the ECB.Journal of International Money and Finance, 104: 102167.
- Bernanke (2020)Bernanke, B.S. 2020.The new tools of monetary policy.American Economic Review, 110(4): 943–983.
- Bharathi and Geetha (2017)Bharathi, S.; and Geetha, A. 2017.Sentiment analysis for effective stock market prediction.International Journal of Intelligent Engineering and Systems, 10(3): 146–154.
- Blinder (2018)Blinder, A.S. 2018.Through a crystal ball darkly: The future of monetary policy communication.In AEA Papers and Proceedings, volume 108, 567–571. American Economic Association 2014 Broadway, Suite 305, Nashville, TN 37203.
- Blinder etal. (2008)Blinder, A.S.; Ehrmann, M.; Fratzscher, M.; DeHaan, J.; and Jansen, D.-J. 2008.Central bank communication and monetary policy: A survey of theory and evidence.Journal of economic literature, 46(4): 910–945.
- Boguth, Grégoire, and Martineau (2019)Boguth, O.; Grégoire, V.; and Martineau, C. 2019.Shaping expectations and coordinating attention: The unintended consequences of FOMC press conferences.Journal of Financial and Quantitative Analysis, 54(6): 2327–2353.
- Busso etal. (2008)Busso, C.; Bulut, M.; Lee, C.; Kazemzadeh, A.; Mower, E.; Kim, S.; Chang, J.N.; Lee, S.; and Narayanan, S.S. 2008.IEMOCAP: interactive emotional dyadic motion capture database.Lang. Resour. Evaluation, 42(4): 335–359.
- Coenen etal. (2017)Coenen, G.; Ehrmann, M.; Gaballo, G.; Hoffmann, P.; Nakov, A.; Nardelli, S.; Persson, E.; and Strasser, G. 2017.Communication of monetary policy in unconventional times.
- Ehrmann and Fratzscher (2007)Ehrmann, M.; and Fratzscher, M. 2007.The timing of central bank communication.European Journal of Political Economy, 23(1): 124–145.
- Frunza (2020)Frunza, O. 2020.Information extraction from federal open market committee statements.In Proceedings of the 1st Joint Workshop on Financial Narrative Processing and MultiLing Financial Summarisation, 195–203.
- Garcia, Hu, and Rohrer (2023)Garcia, D.; Hu, X.; and Rohrer, M. 2023.The colour of finance words.Journal of Financial Economics, 147(3): 525–549.
- Gorodnichenko, Pham, and Talavera (2023)Gorodnichenko, Y.; Pham, T.; and Talavera, O. 2023.The Voice of Monetary Policy.American Economic Review, 113(2): 548–84.
- Gu etal. (2022)Gu, Y.; Tinn, R.; Cheng, H.; Lucas, M.; Usuyama, N.; Liu, X.; Naumann, T.; Gao, J.; and Poon, H. 2022.Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing.ACM Trans. Comput. Heal., 3(1): 2:1–2:23.
- Gururangan etal. (2020)Gururangan, S.; Marasovic, A.; Swayamdipta, S.; Lo, K.; Beltagy, I.; Downey, D.; and Smith, N.A. 2020.Don’t Stop Pretraining: Adapt Language Models to Domains and Tasks.In ACL, 8342–8360. Association for Computational Linguistics.
- Hansen and McMahon (2016)Hansen, S.; and McMahon, M. 2016.Shocking language: Understanding the macroeconomic effects of central bank communication.Journal of International Economics, 99: S114–S133.
- Jordà (2005)Jordà, Ò. 2005.Estimation and inference of impulse responses by local projections.American economic review, 95(1): 161–182.
- Kohn and Sack (2003)Kohn, D.L.; and Sack, B.P. 2003.Central bank talk: does it matter and why?
- Labonte and Makinen (2008)Labonte, M.; and Makinen, G.E. 2008.Monetary policy and the Federal Reserve: Current policy and conditions.Congressional Research Service, Library of Congress.
- Liu etal. (2019)Liu, Y.; Ott, M.; Goyal, N.; Du, J.; Joshi, M.; Chen, D.; Levy, O.; Lewis, M.; Zettlemoyer, L.; and Stoyanov, V. 2019.RoBERTa: A Robustly Optimized BERT Pretraining Approach.CoRR, abs/1907.11692.
- Loughran and Mcdonald (2010)Loughran, T.; and Mcdonald, B. 2010.When is a Liability not a Liability? Textual Analysis, Dictionaries, and 10-Ks.Behavioral & Experimental Accounting eJournal.
- Mathur etal. (2022)Mathur, P.; Neerkaje, A.T.; Chhibber, M.; Sawhney, R.; Guo, F.; Dernoncourt, F.; Dutta, S.; and Manocha, D. 2022.MONOPOLY: Financial Prediction from MONetary POLicY Conference Videos Using Multimodal Cues.In ACM Multimedia, 2276–2285. ACM.
- Matsui, Ren, and Ferrara (2021)Matsui, A.; Ren, X.; and Ferrara, E. 2021.Using word embedding to reveal monetary policy explanation changes.In Proceedings of the Third Workshop on Economics and Natural Language Processing, 56–61.
- Narain and Sangani (2023)Narain, N.; and Sangani, K. 2023.The Market Impact of Fed Communications: The Role of the Press Conference.Available at SSRN 4354333.
- Nasukawa and Yi (2003)Nasukawa, T.; and Yi, J. 2003.Sentiment analysis: capturing favorability using natural language processing.In K-CAP, 70–77. ACM.
- Neuhierl and Weber (2019)Neuhierl, A.; and Weber, M. 2019.Monetary policy communication, policy slope, and the stock market.Journal of Monetary Economics, 108: 140–155.
- Rozkrut etal. (2007)Rozkrut, M.; Rybiński, K.; Sztaba, L.; and Szwaja, R. 2007.Quest for central bank communication: Does it pay to be “talkative”?European Journal of Political Economy, 23(1): 176–206.
- Shah, Paturi, and Chava (2023)Shah, A.; Paturi, S.; and Chava, S. 2023.Trillion Dollar Words: A New Financial Dataset, Task & Market Analysis.In ACL (1), 6664–6679. Association for Computational Linguistics.
- Singer-Vine and The pdfplumber contributors (2023)Singer-Vine, J.; and The pdfplumber contributors. 2023.pdfplumber.
- Tsukioka and Yamasaki (2020)Tsukioka, Y.; and Yamasaki, T. 2020.The tone of the beige book and the pre-fomc announcement drift.Available at SSRN 3306011.
- Yang, Uy, and Huang (2020)Yang, Y.; Uy, M. C.S.; and Huang, A. 2020.FinBERT: A Pretrained Language Model for Financial Communications.CoRR, abs/2006.08097.
- Zirn, Meusel, and Stuckenschmidt (2015)Zirn, C.; Meusel, R.; and Stuckenschmidt, H. 2015.Lost in Discussion? Tracking Opinion Groups in Complex Political Discussions by the Example of the FOMC Meeting Transcriptions.In RANLP, 747–753. RANLP 2015 Organising Committee / ACL.