[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122694-en":3,"doc-seo-122694-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},122694,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",8,"Research & Report","Forecasting inflation using disaggregates and machine learning","This paper evaluates forecasting methods for inflation, emphasizing aggregation of disaggregated forecasts within the bottom-up approach. Using Brazil as an application, the study tests multiple disaggregation levels and combines traditional time-series methods with linear and nonlinear machine learning models to handle a larger set of predictors. For many horizons, aggregated disaggregated forecasts match survey expectations and direct aggregate models. Machine learning improves predictive accuracy overall, with especially strong results for disaggregates, notably during volatile periods after COVID-19.","Forecasting inflation using disaggregates and machine learning  \nGilberto Boaretto  \nDept. of Economics, PUC-Rio, Brazil  \n[gilbertoboaretto@hotmail.com](gilbertoboaretto@hotmail.com)  \nMarcelo C. Medeiros  \nDept. of Economics, UIUC, US  \n[marcelom@illinois.edu](marcelom@illinois.edu)  \narXiv :2308 . 11173v1 [ econ .EM] 22 Aug 2023  \nThis draft: August 23, 2023  \nAbstract  \nThis paper examines the effectiveness of several forecasting methods for predicting inflation, focusing on aggregating disaggregated forecasts – also known in the literature as the bottom-up approach. Taking the Brazilian case as an application, we consider different disaggregation levels for inflation and employ a range of traditional time series techniques as well as linear and nonlinear machine learning (ML) models to deal with a larger number of predictors. For many forecast horizons, the aggregation of disaggregated forecasts performs just as well survey-based expectations and models that generate forecasts using the aggregate directly. Overall, ML methods outperform traditional time series models in predictive accuracy, with outstanding performance in forecasting disaggregates. Our results reinforce the benefits of using models in a data-rich environment for inflation forecasting, including aggregating disaggregated forecasts from ML techniques, mainly during volatile periods. Starting from the COVID-19 pandemic, the random forest model based on both aggregate and disaggregated inflation achieves remarkable predictive performance at intermediate and longer horizons.  \nKeywords: inflation forecasting; disaggregated inflation; bottom-up approach; data-rich environment; machine learning.  \nJEL Codes: C22, C38, C52, C53, C55, E37 .  \nAcknowledgements: Boaretto thanks PUC-Rio, CAPES, CNPq, and FAPERJ for financial support. M˜edeiros thanks CNPq and CAPES for financial support. We thank Gustavo Arajo, Juliano As˜sunc¸ao, Daniel Coutinho, Marcelo Fernandes, Eduardo Freitas, Cleomar Gomes, Fbio Gomes,  \nJoao Victor Issler, Guilherme Kira, Mrcio Laurini, and Gabriel Vasconcelos for helpful comments as well as all seminar participants of the Federal University of Uberlndia Webinar, PUC-Rio Graduate Workshop, XXII Meetin˜g of the Brazilian˜ Society of Finance (SBFin), Depep/BCB Webinar,  \nIta Webinar, University of Sao Paulo at Ribeirao Preto Seminar, and 44th Meeting of the Brazilian Econometric Society (SBE) for numerous comments and suggestions.  \nContents  \n1 Introduction 3  \n2 Forecasting methodology 6  \n2.1 “Traditional” inflation forecasting ............................... 6  \n2.2 Aggregation of disaggregated forecasts ........................... 6  \n2.3 Direct forecasting approach and expanding window scheme ............... 7  \n3 Models and forecast evaluation 8  \n3.1 Models .............................................. 8  \n3.1.1 Benchmarks ....................................... 8  \n3.1.2 Shrinkage-based models ................................ 9  \n3.1.3 Factor models ...................................... 10  \n3.1.4 FarmPredict ....................................... 11  \n3.1.5 Complete subset regression (CSR) .......................... 11  \n3.1.6 Random forest (RF) ................................... 11  \n3.2 Model combinations via average of forecasts ........................ 12  \n3.3 Evaluation: metrics and test .................................. 13  \n4 Data and setup 13  \n5 Results 14  \n5.1 Entire period: forecasts from January 2014 to June 2022 .................. 14  \n5.2 Forecasts before and after of COVID-19 pandemic ..................... 17  \n5.3 Forecast of disaggregates and variable selection ...................... 21  \n5.3.1 Disaggregation into BCB categories ......................... 21  \n5.3.2 Disaggregation into IBGE groups ........................... 26  \n5.3.3 Disaggregation into IBGE subgroups ........................ 26  \n5.3.4 Further remarks ..................................... 27  \n6 Conclusion 28  \nA Groups and subgroups of the I","cbCaifsG5jsxa9FL","https://ap.wps.com/l/cbCaifsG5jsxa9FL","pdf",769577,1,44,"English","en",105,"# Introduction\n# Forecasting methodology\n## Traditional inflation forecasting\n## Aggregation of disaggregated forecasts\n## Direct forecasting approach and expanding window scheme\n# Models and forecast evaluation\n## Models\n## Model combinations via average of forecasts\n## Evaluation: metrics and test\n# Data and setup\n# Results\n## Entire period: forecasts from January 2014 to June 2022\n## Forecasts before and after of COVID-19 pandemic\n## Forecast of disaggregates and variable selection\n# Conclusion","[{\"question\":\"What is the bottom-up approach used in this study for inflation forecasting?\",\"answer\":\"The paper focuses on forecasting inflation components separately and then aggregating the disaggregated forecasts to form an overall inflation forecast.\"},{\"question\":\"How do machine learning models compare with traditional time-series methods?\",\"answer\":\"Across many horizons, the results show that machine learning models deliver better predictive accuracy than traditional time-series benchmarks, with particularly strong performance for forecasting disaggregates.\"},{\"question\":\"What role does COVID-19 play in the forecasting results?\",\"answer\":\"Starting from the COVID-19 pandemic, the study reports notable predictive improvements, especially using a random forest model that leverages both aggregate and disaggregated inflation for intermediate and longer horizons.\"}]","Forecasting inflation using disaggregates and machine learning | PDF",1785812295,111,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"forecasting-inflation-using-disaggregates-and-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/forecasting-inflation-using-disaggregates-and-machine-learning/122694/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is the bottom-up approach used in this study for inflation forecasting?","Question",{"text":75,"@type":76},"The paper focuses on forecasting inflation components separately and then aggregating the disaggregated forecasts to form an overall inflation forecast.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How do machine learning models compare with traditional time-series methods?",{"text":80,"@type":76},"Across many horizons, the results show that machine learning models deliver better predictive accuracy than traditional time-series benchmarks, with particularly strong performance for forecasting disaggregates.",{"name":82,"@type":73,"acceptedAnswer":83},"What role does COVID-19 play in the forecasting results?",{"text":84,"@type":76},"Starting from the COVID-19 pandemic, the study reports notable predictive improvements, especially using a random forest model that leverages both aggregate and disaggregated inflation for intermediate and longer horizons.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]