Comparative computer simulation and empirical analysis of MIDAS and artificial neural network-UMIDAS models for short- and long-term US GDP forecasting
Samir K H. Safi, Olajide Idris Sanusi, Afreen Arif
{"title":"Comparative computer simulation and empirical analysis of MIDAS and artificial neural network-UMIDAS models for short- and long-term US GDP forecasting","authors":"Samir K H. Safi, Olajide Idris Sanusi, Afreen Arif","doi":"10.1108/cr-09-2023-0238","DOIUrl":null,"url":null,"abstract":"<h3>Purpose</h3>\n<p>This study aims to evaluate linear mixed data sampling (MIDAS), nonlinear artificial neural networks (ANNs) and a hybrid approach for exploiting high-frequency information to improve low-frequency gross domestic product (GDP) forecasting. Their capabilities are assessed through direct forecasting comparisons.</p><!--/ Abstract__block -->\n<h3>Design/methodology/approach</h3>\n<p>This study compares quarterly GDP forecasts from unrestricted MIDAS (UMIDAS), standalone ANN and ANN-enhanced MIDAS models using five monthly predictors. Rigorous empirical analysis of recent US data is supplemented by Monte Carlo simulations to validate findings.</p><!--/ Abstract__block -->\n<h3>Findings</h3>\n<p>The empirical results and simulations demonstrate that the hybrid ANN-MIDAS performs best for short-term predictions, whereas UMIDAS is more robust for long-term forecasts. The integration of ANNs into MIDAS provides modeling flexibility and accuracy gains for near-term forecasts.</p><!--/ Abstract__block -->\n<h3>Research limitations/implications</h3>\n<p>The model comparisons are limited to five selected monthly indicators. Expanding the variables and alternative data processing techniques may reveal further insights. Longer analysis horizons could identify structural breaks in relationships.</p><!--/ Abstract__block -->\n<h3>Practical implications</h3>\n<p>The findings guide researchers and policymakers in leveraging mixed frequencies amidst data complexity. Appropriate modeling choices based on context and forecast horizon can maximize accuracy.</p><!--/ Abstract__block -->\n<h3>Social implications</h3>\n<p>Enhanced GDP forecasting supports improved policy and business decisions, benefiting economic performance and societal welfare. More accurate predictions build stakeholder confidence and trust in statistics underlying critical choices.</p><!--/ Abstract__block -->\n<h3>Originality/value</h3>\n<p>This direct forecasting comparison offers unique large-scale simulation evidence on harnessing mixed frequencies with leading statistical and machine learning techniques. The results elucidate their complementarity for short-term versus long-term modeling.</p><!--/ Abstract__block -->","PeriodicalId":46521,"journal":{"name":"Competitiveness Review","volume":"9 1","pages":""},"PeriodicalIF":2.9000,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Competitiveness Review","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1108/cr-09-2023-0238","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"BUSINESS","Score":null,"Total":0}
引用次数: 0
Abstract
Purpose
This study aims to evaluate linear mixed data sampling (MIDAS), nonlinear artificial neural networks (ANNs) and a hybrid approach for exploiting high-frequency information to improve low-frequency gross domestic product (GDP) forecasting. Their capabilities are assessed through direct forecasting comparisons.
Design/methodology/approach
This study compares quarterly GDP forecasts from unrestricted MIDAS (UMIDAS), standalone ANN and ANN-enhanced MIDAS models using five monthly predictors. Rigorous empirical analysis of recent US data is supplemented by Monte Carlo simulations to validate findings.
Findings
The empirical results and simulations demonstrate that the hybrid ANN-MIDAS performs best for short-term predictions, whereas UMIDAS is more robust for long-term forecasts. The integration of ANNs into MIDAS provides modeling flexibility and accuracy gains for near-term forecasts.
Research limitations/implications
The model comparisons are limited to five selected monthly indicators. Expanding the variables and alternative data processing techniques may reveal further insights. Longer analysis horizons could identify structural breaks in relationships.
Practical implications
The findings guide researchers and policymakers in leveraging mixed frequencies amidst data complexity. Appropriate modeling choices based on context and forecast horizon can maximize accuracy.
Social implications
Enhanced GDP forecasting supports improved policy and business decisions, benefiting economic performance and societal welfare. More accurate predictions build stakeholder confidence and trust in statistics underlying critical choices.
Originality/value
This direct forecasting comparison offers unique large-scale simulation evidence on harnessing mixed frequencies with leading statistical and machine learning techniques. The results elucidate their complementarity for short-term versus long-term modeling.
期刊介绍:
The following list indicates the key issues in the Competitiveness Review. We invite papers on these and related topics. Special issues of the Review will collect papers on specific topics selected by the editors. Definition/conceptual framework of competitiveness Competitiveness diagnostics and rankings Competitiveness and economic outcomes Specific dimensions of competitiveness Competitiveness and endowments Competitiveness and economic development Location and business strategy International business and the role of MNCs Innovation and innovative capacity Clusters and cluster initiatives Institutions for competitiveness Public policy (e.g., innovation, cluster development, regional development) The Competitiveness Review aims to publish high quality papers directed at scholars, government institutions, businesses and practitioners. It appears in collaboration with key academic and professional groups in the field of competitiveness analysis and policy, including the Microeconomics of Competitiveness (MOC) network and The Competitiveness Institute (TCI) practitioner network for competitiveness, clusters and innovation.