|
on Computational Economics |
| By: | Maheronnaghsh, Mohammad Javad; Gheidi, Mohammad Mahdi; Younesi, Abolfazl; Fazli, MohammadAmin |
| Abstract: | In the dynamic world of financial markets, accurate price predictions are essential for informed decision-making. This research proposal outlines a comprehensive study aimed at forecasting stock and currency prices using state-of-the-art Machine Learning (ML) techniques. By delving into the intricacies of models such as Transformers, LSTM, Simple RNN, NHits, and NBeats, we seek to contribute to the realm of financial forecasting, offering valuable insights for investors, financial analysts, and researchers. This article provides an in-depth overview of our methodology, data collection process, model implementations, evaluation metrics, and potential applications of our research findings. The research indicates that NBeats and NHits models exhibit superior performance in financial forecasting tasks, especially with limited data, while Transformers require more data to reach full potential. Our findings offer insights into the strengths of different ML techniques for financial prediction, highlighting specialized models like NBeats and NHits as top performers - thus informing model selection for real-world applications. To enhance readability, all acronyms used in the paper are defined below: ML: Machine Learning LSTM: Long Short-Term Memory RNN: Recurrent Neural Network NHits: Neural Hierarchical Interpolation for Time Series Forecasting NBeats: Neural Basis Expansion Analysis for Time Series ARIMA: Autoregressive Integrated Moving Average GARCH: Generalized Autoregressive Conditional Heteroskedasticity SVMs: Support Vector Machines CNNs: Convolutional Neural Networks MSE: Mean Squared Error MAE: Mean Absolute Error RMSE: Recurrent Mean Squared Error API: Application Programming Interface F1-score: F1 Score GRU: Gated Recurrent Unit yfinance: Yahoo Finance (a Python library for fetching financial data) |
| Date: | 2023–09–30 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:dzp26_v1 |
| By: | Andrés Alonso-Robisco (BANCO DE ESPAÑA); Andrés Azqueta-Gavaldón (BANCO DE ESPAÑA); José Manuel Carbó (BANCO DE ESPAÑA); José Luis González (BANCO DE ESPAÑA); Ana Isabel Hernáez (BANCO DE ESPAÑA); José Luis Herrera (BANCO DE ESPAÑA); Jorge Quintana (BANCO DE ESPAÑA); Javier Tarancón (BANCO DE ESPAÑA) |
| Abstract: | New technologies have made available a vast amount of new data in the form of text, recording an exponentially increasing share of human and corporate behavior. For financial supervisors, the information encoded in text is a valuable complement to the more traditional balance sheet data typically used to track the soundness of financial institutions. In this study, we exploit several natural language processing (NLP) techniques as well as network analysis to detect anomalies in the Spanish corporate system, identifying both idiosyncratic and systemic risks. We use sentiment analysis at the corporate level to detect sentiment anomalies for specific corporations (idiosyncratic risks), while employing a wide range of network metrics to monitor systemic risks. In the realm of supervisory technology (SupTech), anomaly detection in sentiment analysis serves as a proactive tool for financial authorities. By continuously monitoring sentiment trends, SupTech applications can provide early warnings of potential financial distress or systemic risks. |
| Keywords: | suptech, natural language processing, machine learning, network analysis, sentiment |
| JEL: | C63 D81 G21 |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:bde:opaper:2504 |
| By: | Pletcher, Scott Nicholas |
| Abstract: | For many organizations, artificial intelligence and its subsets of machine learning and deep learning hold great potential for improving efficiency, creating new capabilities and launching new business models. Accordingly, many organizations are attempting to harness these technologies through prototyping and pilot projects. However, many organizations struggle to move past the pilot phase, despite heavy investment in time, data infrastructure and training. In their book Strategic Doing, Morrison et al. (2019) provide a framework to help organizations brainstorm, organize and launch innovation using ten skills of agile leadership. A specific step in the described approach is to Start Slowly to Go Fast. This simple statement holds some deep implications, with many of the principles contained within that philosophy shown to improve innovation outcomes. This paper will examine some of those principles in the context of AI projects. |
| Date: | 2023–09–01 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:8jqzu_v1 |
| By: | Kimihiko Izawa (Bank of Japan); Ikuo Kamei (Bank of Japan); Nao Shibata (Bank of Japan); Yusuke Takahashi (Bank of Japan); Shunichi Yoneyama (Bank of Japan) |
| Abstract: | This paper examines whether textual data analysis using Large Language Models (LLMs) can be applied to assessing economic activity and prices in light of the rapid development of LLMs in recent years. LLMs have advantages in that there are a wide range of models available for use without large initial costs and that these models, which have already acquired basic knowledge of language, can analyze any topic or text and are beginning to be used in economic analysis more widely, including those of central banks. This paper, as an example, attempts to use LLMs to analyze recent wage and price developments in Japan using comments from the Cabinet Office's Economy Watchers Survey. The results suggest that the cause of increasing selling prices is gradually shifting from raw material costs to labor costs. |
| Date: | 2025–03–24 |
| URL: | https://d.repec.org/n?u=RePEc:boj:bojrev:rev25e05 |
| By: | Vasilios Plakandaras (Department of Economics, Democritus University of Thrace, Komotini, Greece); Rangan Gupta (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Qiang Ji (Institutes of Science and Development, Chinese Academy of Sciences, Beijing, China; School of Public Policy and Management, University of Chinese Academy of Sciences, Beijing, 100049, China) |
| Abstract: | The study investigates systemic financial risk in global markets, attributing it to geopolitical instability, climate risks, and economic uncertainties. Utilizing a state-of-the-art machine learning heterogeneous panel regression framework capable of capturing cross-sectional dependencies and nonlinear patterns, we examine financial stress across multiple economies, including China, the U.S., the U.K., and ten EU nations. Through extensive out-of-sample rolling window analysis, we show that while geopolitical uncertainty enhances short-term predictions, long-term risk forecasting is better achieved using financial and economic data. The study underscores the limitations of conventional regression models in capturing financial risk dynamics and suggests that machine learning-based panel regressions provide a more nuanced and accurate forecasting tool. The findings bear significant policy implications, highlighting the necessity for regulatory bodies to reassess risk frameworks and the role of climate-related disclosures in financial markets. |
| Keywords: | Systemic financial risk, machine learning, forecasting, climate risk, geopolitical risk |
| JEL: | C45 C58 G17 |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:pre:wpaper:202511 |
| By: | Bayona, Pamela |
| Abstract: | This paper explores the application of Natural Language Processing (NLP) techniques to automate Harmonized System (HS) tariff line transposition, employing a three-stage process: unique 1:1 tariff code matching (Round 1), exact description matching (Round 2), and "smart" description matching (Round 3) using Artificial Intelligence (AI) and lexical similarity methods paired with harmonized 6- digit concordance and cosine similarity. Similarity is calculated using either Term Frequency Inverse Document Frequency (TF-IDF) vectors or Sentence-BERT (SBERT) embeddings, comparing two scenarios: a straightforward case (Economy A) with standardized descriptions, and a complex case (Economy B), with more detailed technical descriptions. Results indicate that automated HS transposition can significantly augment the efficiency of traditionally manual methods, reducing processing time from two to three weeks to approximately half a day (up to 30 times faster). The overall accuracy rate is 99.6% for the simpler scenario and 98.8% for the complex one, for a standard set of approximately 10, 000 HS codes. While non-AI techniques cover most of the accurate matches, AI-based Round 3 techniques address cases requiring the most manual effort. SBERT generally outperforms TF-IDF, however including subheadings tends to reduce its accuracy. In certain cases, particularly for highly technical tariffs, TF-IDF's straightforward approach provides an advantage over SBERT. Overall, NLP techniques hold significant potential for improving HS transposition methods and facilitating the development of richer tariffs and trade datasets to enable more in-depth analyses. Future research should focus on refining these techniques across diverse datasets to optimize their broader application in tariff and trade data analysis. |
| Keywords: | Harmonized System, tariff line, HS transposition, correlation tables, concordance, natural language processing |
| JEL: | F10 F13 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:wtowps:314422 |
| By: | Mathias, Lea |
| Abstract: | Multi-level Fusion in Deep Convolutional Networks for Enhanced Image Analysis |
| Date: | 2023–08–10 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:58h7t_v1 |
| By: | Mamalis, Marios; Kalampokis, Evangelos; Karamanou, Areti; Brimos, Petros; Tarabanis, Konstantinos |
| Abstract: | Large language models possess tremendous natural language understanding and generation abilities. However, they often lack the ability to discern between fact and fiction, leading to factually incorrect responses. Open Government Data are repositories of, often times linked, information that is freely available to everyone. By combining these two technologies in a proof of concept designed application utilizing the GPT3.5 OpenAI model and the Scottish open statistics portal, we show that not only is it possible to augment the large language model's factuality of responses, but also propose a novel way to effectively access and retrieve statistical information from the data portal just through natural language querying. We anticipate that this paper will trigger a discussion regarding the transformation of Open Government Portals through large language models. |
| Date: | 2023–10–24 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:9b35z_v1 |
| By: | Phoebe Koundouri; Angelos Alamanos; Giannis Arampatzidis; Stathis Devves; Jeffrey D Sachs |
| Abstract: | As global commitments to decarbonization intensify, energy-emission models are becoming increasingly vital for policymaking, offering data-driven insights to evaluate the feasibility and impact of climate strategies. These models help governments design evidence-based policies, assess mitigation pathways, and ensure alignment with national and international targets, such as the Paris Agreement and the EU Green Deal. Researchers often spend a lot of time considering their modelling choices to develop the best possible tools in terms of data-requirements, accuracy, computational demand, while there is always a 'debate' of complexity versus explicability and ready-to-use models for policymaking. Especially for energy-emissions models, given their increasing policy-relevance, and the need to provide insights fast for short-term policies (e.g. 2030, or 2050 net-zero goals), such considerations become increasingly pressing. In this paper, we present two different versions of the same energy-emissions model, and we run them for the same study area, planning horizon, and scenario analysis. The two versions differ only in how they approach complexity: Version1 is a more 'detailed', complex model, while Version2 is a 'simpler' and less data-hungry one. A set of evaluation criteria was then used to qualitatively compare these two versions, based on modelling- and policymaking-related considerations, debating modelers' and policymakers' expectations and preferences. We reflect on best modelling practices, discuss different goal-dependent approaches, providing useful guidance for modelers and policymakers. |
| Keywords: | Energy-emissions modelling, Decarbonization pathways, Model development, LEAP, Models to policy. |
| Date: | 2025–03–28 |
| URL: | https://d.repec.org/n?u=RePEc:aue:wpaper:2528 |
| By: | Lisandra Flach; Andreas Baur |
| Abstract: | Key MessagesUS President Donald Trump has recently announced 25% tariffs on US imports from Canada and Mexico. A simulation analysis using a quantitative framework shows that Trump’s tariffs would hit the manufacturing sector of the US’s North American neighbors particularly hard.In the event that Mexico and Canada impose symmetric retaliatory tariffs, all sectors of the economy (services, agriculture, and manufacturing) would incur permanent value-added losses.In Mexico and Canada, manufacturing has the largest decline in value added, 13% and 14%, respectively.For the US economy, agriculture has the largest decline in value added (-2.39%), but other sectors of the economy also incur permanent losses.In the event of retaliation, Canada would have to expect a long-term permanent decline in total exports of up to 28%, while Mexico could see a drop of more than 35% and the US of 22%. |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:ces:econpb:_70 |
| By: | Sonali Chowdhry |
| Abstract: | US trade policy has taken a sharp turn away from multilateralism, with sweeping new tariffs posing a serious threat to global supply chains. As the US remains the EU’s largest export market for goods, these measures carry significant repercussions for the bloc. Exports to the US are heavily reliant on a small number of companies and high-value business relationships—making the EU particularly vulnerable to targeted trade measures. In Germany, the top ten business relationships alone account for a fifth of maritime exports to the US. Intra-company trade also plays a crucial role: One quarter of automotive exports from Germany to the US is between business entities with clear common ownership. Simulations further suggest that a transatlantic tariff conflict would halve EU exports to the US and trigger widespread production losses, with Germany’s GDP contracting by approximately 0.33% in the long term. To limit these economic damages and build long-term resilience, the EU should accelerate its export diversification by deepening trade ties with Free Trade Agreement partners and enhancing integration within the single market. |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:diw:diwfoc:11en |