|
on Computational Economics |
|
Issue of 2024‒04‒22
eighteen papers chosen by |
| By: | Maheronnaghsh, Mohammad Javad; Gheidi, Mohammad Mahdi; 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: | http://d.repec.org/n?u=RePEc:osf:osfxxx:dzp26&r=cmp |
| By: | Thanos Konstantinidis; Giorgos Iacovides; Mingxue Xu; Tony G. Constantinides; Danilo Mandic |
| Abstract: | There are multiple sources of financial news online which influence market movements and trader's decisions. This highlights the need for accurate sentiment analysis, in addition to having appropriate algorithmic trading techniques, to arrive at better informed trading decisions. Standard lexicon based sentiment approaches have demonstrated their power in aiding financial decisions. However, they are known to suffer from issues related to context sensitivity and word ordering. Large Language Models (LLMs) can also be used in this context, but they are not finance-specific and tend to require significant computational resources. To facilitate a finance specific LLM framework, we introduce a novel approach based on the Llama 2 7B foundational model, in order to benefit from its generative nature and comprehensive language manipulation. This is achieved by fine-tuning the Llama2 7B model on a small portion of supervised financial sentiment analysis data, so as to jointly handle the complexities of financial lexicon and context, and further equipping it with a neural network based decision mechanism. Such a generator-classifier scheme, referred to as FinLlama, is trained not only to classify the sentiment valence but also quantify its strength, thus offering traders a nuanced insight into financial news articles. Complementing this, the implementation of parameter-efficient fine-tuning through LoRA optimises trainable parameters, thus minimising computational and memory requirements, without sacrificing accuracy. Simulation results demonstrate the ability of the proposed FinLlama to provide a framework for enhanced portfolio management decisions and increased market returns. These results underpin the ability of FinLlama to construct high-return portfolios which exhibit enhanced resilience, even during volatile periods and unpredictable market events. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.12285&r=cmp |
| By: | Moseli Mots'oehli; Anton Nikolaev; Wawan B. IGede; John Lynham; Peter J. Mous; Peter Sadowski |
| Abstract: | Fish stock assessment often involves manual fish counting by taxonomy specialists, which is both time-consuming and costly. We propose an automated computer vision system that performs both taxonomic classification and fish size estimation from images taken with a low-cost digital camera. The system first performs object detection and segmentation using a Mask R-CNN to identify individual fish from images containing multiple fish, possibly consisting of different species. Then each fish species is classified and the predicted length using separate machine learning models. These models are trained on a dataset of 50, 000 hand-annotated images containing 163 different fish species, ranging in length from 10cm to 250cm. Evaluated on held-out test data, our system achieves a $92\%$ intersection over union on the fish segmentation task, a $89\%$ top-1 classification accuracy on single fish species classification, and a $2.3$~cm mean error on the fish length estimation task. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.10916&r=cmp |
| By: | Cecilia Ying; Stephen Thomas |
| Abstract: | In an effort to improve the accuracy of credit lending decisions, many financial intuitions are now using predictions from machine learning models. While such predictions enjoy many advantages, recent research has shown that the predictions have the potential to be biased and unfair towards certain subgroups of the population. To combat this, several techniques have been introduced to help remove the bias and improve the overall fairness of the predictions. We introduce a new fairness technique, called \textit{Subgroup Threshold Optimizer} (\textit{STO}), that does not require any alternations to the input training data nor does it require any changes to the underlying machine learning algorithm, and thus can be used with any existing machine learning pipeline. STO works by optimizing the classification thresholds for individual subgroups in order to minimize the overall discrimination score between them. Our experiments on a real-world credit lending dataset show that STO can reduce gender discrimination by over 90\%. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.10652&r=cmp |
| By: | Jonathan Fuhr (School of Business and Economics, University of T\"ubingen); Philipp Berens (Hertie Institute for AI in Brain Health, University of T\"ubingen); Dominik Papies (School of Business and Economics, University of T\"ubingen) |
| Abstract: | The estimation of causal effects with observational data continues to be a very active research area. In recent years, researchers have developed new frameworks which use machine learning to relax classical assumptions necessary for the estimation of causal effects. In this paper, we review one of the most prominent methods - "double/debiased machine learning" (DML) - and empirically evaluate it by comparing its performance on simulated data relative to more traditional statistical methods, before applying it to real-world data. Our findings indicate that the application of a suitably flexible machine learning algorithm within DML improves the adjustment for various nonlinear confounding relationships. This advantage enables a departure from traditional functional form assumptions typically necessary in causal effect estimation. However, we demonstrate that the method continues to critically depend on standard assumptions about causal structure and identification. When estimating the effects of air pollution on housing prices in our application, we find that DML estimates are consistently larger than estimates of less flexible methods. From our overall results, we provide actionable recommendations for specific choices researchers must make when applying DML in practice. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.14385&r=cmp |
| By: | N. Meltem Daysal (University of Copenhagen, CEBI, CESIfo, IZA); Sendhil Mullainathan (University of Chicago Booth School of Business); Ziad Obermeyer (University of California, Berkeley); Suproteem K. Sarkar (Harvard University); Mircea Trandafir (The Rockwool Foundation Research Unit) |
| Abstract: | We consider the health effects of “precision†screening policies for cancer guided by algorithms. We show that machine learning models that predict breast cancer from health claims data outperform models based on just age and established risk factors. We estimate that screening women with high predicted risk of invasive tumors would reduce the long-run incidence of later-stage tumors by 40%. Screening high-risk women would also lead to half the rate of cancer overdiagnosis that screening low-risk women would. We show that these results depend crucially on the machine learning model’s prediction target. A model trained to predict positive mammography results leads to policies with weaker health effects and higher rates of overdiagnosis than a model trained to predict invasive tumors. |
| Keywords: | breast cancer, precision screening, predictive modeling, machine leaning, health policy |
| JEL: | I12 I18 J16 C55 |
| Date: | 2022–12–17 |
| URL: | http://d.repec.org/n?u=RePEc:kud:kucebi:2224&r=cmp |
| By: | Mueller, H.; Rauh, C.; Seimon, B. |
| Abstract: | This article provides a structured description of openly available news topics and forecasts for armed conflict at the national and grid cell level starting January 2010. The news topics as well as the forecasts are updated monthly at conflictforecast.org and provide coverage for more than 170 countries and about 65, 000 grid cells of size 55x55km worldwide. The forecasts rely on Natural Language Processing (NLP) and machine learning techniques to leverage a large corpus of newspaper text for predicting sudden onsets of violence in peaceful countries. Our goals are to: a) support conflict prevention efforts by making our risk forecasts available to practitioners and research teams worldwide, b) facilitate additional research that can utilise risk forecasts for causal identification, and to c) provide an overview of the news landscape. |
| Keywords: | Civil War, Conflict, Forecasting, Machine Learning, News Topics, Random Forest, Topic Models |
| Date: | 2024–02–02 |
| URL: | http://d.repec.org/n?u=RePEc:cam:camjip:2402&r=cmp |
| By: | Lucas javaudin; André de Palma (Université de Cergy-Pontoise, THEMA) |
| Abstract: | Transport simulators can be used to compute the equilibrium between transporta- tion demand and supply within complex transportation systems. However, despite their theoretical foundations, there is a lack of comparative analysis between simula- tor results and theoretical models in the literature. In this paper, we bridge this gap by introducing METROPOLIS2, a novel mesoscopic transport simulator capable of simulating agents’ travel decisions (including mode, departure-time, and route choice), based on discrete-choice theory within a dynamic, continuous-time framework. We demonstrate METROPOLIS2’s functionality through its application to the single-road bottleneck model and validate its ability to replicate analytical results. Furthermore, we provide a comprehensive overview of METROPOLIS2 in large-scale scenarios. Fi- nally, we compare METROPOLIS2’s results with those of the original METROPOLIS1 simulator in a simulation of Paris, highlighting its speed and ability to converge to an equilibrium. |
| Keywords: | transport simulation; agent-based modeling; bottleneck; dynamic traffic assignment; discrete-choice models |
| JEL: | C63 R4 |
| Date: | 2024 |
| URL: | http://d.repec.org/n?u=RePEc:ema:worpap:2024-03&r=cmp |
| By: | Bruno de Melo |
| Abstract: | Performance attribution analysis, defined as the process of explaining the drivers of the excess performance of an investment portfolio against a benchmark, stands as a significant aspect of portfolio management and plays a crucial role in the investment decision-making process, particularly within the fund management industry. Rooted in a solid financial and mathematical framework, the importance and methodologies of this analytical technique are extensively documented across numerous academic research papers and books. The integration of large language models (LLMs) and AI agents marks a groundbreaking development in this field. These agents are designed to automate and enhance the performance attribution analysis by accurately calculating and analyzing portfolio performances against benchmarks. In this study, we introduce the application of an AI Agent for a variety of essential performance attribution tasks, including the analysis of performance drivers and utilizing LLMs as calculation engine for multi-level attribution analysis and question-answer (QA) exercises. Leveraging advanced prompt engineering techniques such as Chain-of-Thought (CoT) and Plan and Solve (PS), and employing a standard agent framework from LangChain, the research achieves promising results: it achieves accuracy rates exceeding 93% in analyzing performance drivers, attains 100% in multi-level attribution calculations, and surpasses 84% accuracy in QA exercises that simulate official examination standards. These findings affirm the impactful role of AI agents, prompt engineering and evaluation in advancing portfolio management processes, highlighting a significant advancement in the practical application and evaluation of AI technologies within the domain. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.10482&r=cmp |
| By: | Eli Ben-Michael; D. James Greiner; Melody Huang; Kosuke Imai; Zhichao Jiang; Sooahn Shin |
| Abstract: | The use of Artificial Intelligence (AI) based on data-driven algorithms has become ubiquitous in today's society. Yet, in many cases and especially when stakes are high, humans still make final decisions. The critical question, therefore, is whether AI helps humans make better decisions as compared to a human alone or AI an alone. We introduce a new methodological framework that can be used to answer experimentally this question with no additional assumptions. We measure a decision maker's ability to make correct decisions using standard classification metrics based on the baseline potential outcome. We consider a single-blinded experimental design, in which the provision of AI-generated recommendations is randomized across cases with a human making final decisions. Under this experimental design, we show how to compare the performance of three alternative decision-making systems--human-alone, human-with-AI, and AI-alone. We apply the proposed methodology to the data from our own randomized controlled trial of a pretrial risk assessment instrument. We find that AI recommendations do not improve the classification accuracy of a judge's decision to impose cash bail. Our analysis also shows that AI-alone decisions generally perform worse than human decisions with or without AI assistance. Finally, AI recommendations tend to impose cash bail on non-white arrestees more often than necessary when compared to white arrestees. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.12108&r=cmp |
| By: | Qirui Mi; Zhiyu Zhao; Siyu Xia; Yan Song; Jun Wang; Haifeng Zhang |
| Abstract: | Effective macroeconomic policies play a crucial role in promoting economic growth and social stability. This paper models the optimal macroeconomic policy problem based on the \textit{Stackelberg Mean Field Game} (SMFG), where the government acts as the leader in policy-making, and large-scale households dynamically respond as followers. This modeling method captures the asymmetric dynamic game between the government and large-scale households, and interpretably evaluates the effects of macroeconomic policies based on microfoundations, which is difficult for existing methods to achieve. We also propose a solution for SMFGs, incorporating pre-training on real data and a model-free \textit{Stackelberg mean-field reinforcement learning }(SMFRL) algorithm, which operates independently of prior environmental knowledge and transitions. Our experimental results showcase the superiority of the SMFG method over other economic policies in terms of performance, efficiency-equity tradeoff, and SMFG assumption analysis. This paper significantly contributes to the domain of AI for economics by providing a powerful tool for modeling and solving optimal macroeconomic policies. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.12093&r=cmp |
| By: | Bertin Martens |
| Abstract: | The licensing of training inputs slows down economic growth compared to what it could be with competitive and high-quality GenAI |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:bre:wpaper:node_9853&r=cmp |
| By: | Suss, Joel; Kemeny, Tom; Connor, Dylan S. |
| Abstract: | Wealth inequality has been sharply rising in the United States and across many other high-income countries. Due to a lack of data, we know little about how this trend has unfolded across locations within countries. Examining the subnational geography of wealth is crucial because, from one generation to the next, it shapes the distribution of opportunity, disadvantage, and power across individuals and communities. By employing machine-learning-based imputation to link national historical surveys conducted by the U.S. Federal Reserve to population survey microdata, the data presented in this article addresses this gap. The Geographic Wealth Inequality Database (“GEOWEALTH-US”) provides the first estimates of the level and distribution of wealth at various geographical scales within the United States from 1960 to 2020. The GEOWEALTH-US database enables new lines of investigation into the contribution of spatial wealth disparities to major societal challenges including wealth concentration, income inequality, social mobility, housing unaffordability, and political polarization. |
| JEL: | N0 |
| Date: | 2024–02–28 |
| URL: | http://d.repec.org/n?u=RePEc:ehl:lserod:122377&r=cmp |
| By: | Harashima, Taiji |
| Abstract: | In this paper, I numerically simulate the path of economy in an economic depression. It is not easy to perform a numerical simulation of the path to a steady state if households are assumed to behave by generating rational expectations. It is much easier, however, if households are assumed to behave according to a procedure based on the maximum degree of comfortability (MDC), where MDC indicates the state at which a household feels most comfortable with its combination of income and assets. The results of simulations under the supposition of this alternative procedure indicate that, if households do not strategically consider other households’ behaviors, consumption jumps upwards immediately after the shock. However, if households strategically select a Pareto inefficient path, large amounts of unutilized economic resources are generated, and the unemployment rate can rise to 30% or higher. These results seem to well match actual historical experiences during severe recessions such as the Great Depression and Great Recession. |
| Keywords: | Economic depression; Shock; Simulation; Recession; Unemployment; Unutilized economic resources |
| JEL: | E10 E17 E27 E32 E37 |
| Date: | 2024–03–21 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:120508&r=cmp |
| By: | Allwein, Florian |
| Abstract: | ChatGPT, a chatbot based on a Large Language Model, has become one of the fastest-growing consumer software applications in history. Discussion of the tool and its use cases contains an element of hype as it appears that the technology's capabilities are sometimes exaggerated. Moreover, a critical perspective is often lacking in research and practice. This paper points out some significant downsides, risks and limitations of using ChatGPT, arguing for a critical view of the tool based on ethics, regulations and reflected use. This can be used as a guideline for decisions on whether and how to use ChatGPT, and can inform future research. |
| Keywords: | ChatGPT, AI, Ethics |
| Date: | 2024 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:iubhit:287746&r=cmp |
| By: | Hans Gersbach; Fikri Pitsuwan; Pio Blieske |
| Abstract: | Bug bounty programs, where external agents are invited to search and report vulnerabilities (bugs) in exchange for rewards (bounty), have become a major tool for companies to improve their systems. We suggest augmenting such programs by inserting artificial bugs to increase the incentives to search for real (organic) bugs. Using a model of crowdsearch, we identify the efficiency gains by artificial bugs, and we show that for this, it is sufficient to insert only one artificial bug. Artificial bugs are particularly beneficial, for instance, if the designer places high valuations on finding organic bugs or if the budget for bounty is not sufficiently high. We discuss how to implement artificial bugs and outline their further benefits. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.09484&r=cmp |
| By: | Espíndola, Ernesto; Suárez, José Ignacio |
| Abstract: | In recent decades, rapid technological progress has generated a growing interest in the transformation of the world of work. This concern is based on the potential of emerging technologies to replace tasks and roles traditionally performed by human beings, either partially or entirely. It is, therefore, essential to examine and understand the social, economic, and ethical implications of this process and seek solutions to harness the benefits associated with the automation of production processes and mitigate possible negative impacts. This paper seeks to estimate job automation's probabilities and risks and analyse its potential impacts on labour inclusion in Latin America. To this end, this document implemented a machine learning-based methodology adapted to the specific characteristics of the region using data from PIAAC surveys and household surveys. In this way, the aim is to build a probability vector of job automation adapted to the region. This vector can be reused in any source of information that contains internationally comparable occupational codes, such as household surveys or employment surveys. The study provides novel estimates of labour automation based on Latin American data and analyses the phenomenon in different aspects of labour inclusion and social stratification. The results show that the risks of automation vary among different social groups, which points to the need to build adapted and efficient policies that address the diverse needs that this process imposes. To this end, the document addresses different policy areas to promote effective labour inclusion in an era of rapid advances in intelligent technologies, ensuring that all individuals can access decent employment and so that these inequalities can be addressed effectively. |
| Date: | 2024–03–25 |
| URL: | http://d.repec.org/n?u=RePEc:ecr:col041:69088&r=cmp |
| By: | Anton Korinek; Donghyun Suh |
| Abstract: | We analyze how output and wages behave under different scenarios for technological progress that may culminate in Artificial General Intelligence (AGI), defined as the ability of AI systems to perform all tasks that humans can perform. We assume that human work can be decomposed into atomistic tasks that differ in their complexity. Advances in technology make ever more complex tasks amenable to automation. The effects on wages depend on a race between automation and capital accumulation. If the distribution of task complexity exhibits a sufficiently thick infinite tail, then there is always enough work for humans, and wages may rise forever. By contrast, if the complexity of tasks that humans can perform is bounded and full automation is reached, then wages collapse. But declines may occur even before if large-scale automation outpaces capital accumulation and makes labor too abundant. Automating productivity growth may lead to broad-based gains in the returns to all factors. By contrast, bottlenecks to growth from irreproducible scarce factors may exacerbate the decline in wages. |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2403.12107&r=cmp |