[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-290854-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-290854-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","machine-learning-for-marketing-mix-modeling-improving-accuracy-and-interpretability","Machine Learning for Marketing Mix Modeling - Improving Accuracy and Interpretability","","Bachelor thesis developing a Machine Learning for Marketing Mix Modeling (MMM) approach to improve marketing performance modeling. It replaces linear assumptions between media variables and company sales with models capturing non-linear effects of advertising investment. Accuracy gains are paired with interpretability by combining Generalized Additive Models (GAM) and Gradient Boosting Regressor (GBR) and explaining predictions using SHAP values. The work builds baseline regression models, applies adstock and saturation transformations, optimizes channel investment for maximum sales, and presents results through informative plots.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/machine-learning-for-marketing-mix-modeling-improving-accuracy-and-interpretability/290854/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/machine-learning-for-marketing-mix-modeling-improving-accuracy-and-interpretability/290854.png","ImageObject",300,407,{"name":42,"@type":43},"Skyler","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-29","2026-09-17",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",13,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What is the main objective of the thesis?","Question",{"text":63,"@type":64},"To develop a Machine Learning-based MMM approach that improves accuracy over current sector models that assume linear relationships between media variables and sales.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does the thesis improve interpretability of the marketing models?",{"text":68,"@type":64},"It combines GAM and Gradient Boosting Regressor, then uses SHAP values to provide interpretability for what the model is doing.",{"name":70,"@type":61,"acceptedAnswer":71},"What transformations and modeling baselines are used before final optimization?",{"text":72,"@type":64},"The workflow includes automated data visualization and transformation, baseline linear regression models, and marketing transformations such as adstock and saturation before analyzing results and optimizing investments.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},290854,1789642814,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":140,"language":141,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":12,"update_tm":80,"read_time":145,"preview_page_count":4},2336464648746,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Bachelor in Data Science and Engineering  \nBACHELOR THESIS  \nMachine Learning for Marketing Mix Modeling: Improving Accuracy and Interpretability  \nMarcos Ferreras Arroyo Tutor: Francisco Javier Nogales Martín  \nLeganés, June 2025  \nThis work is licensed under the Creative Commons Attribution-Non CommercialNon Derivatives.  \n2  \nSUMMARY  \nThe main objective of this thesis is to use machine learning algorithms to develop a MMM ( Marketing Mix Modelling) approach. Machine learning allows models to achieve a higher accuracy than the ones being used in the sector at the moment, which assume a linear relationship between the media variables and the company’s overall sales. Furthermore, ML models take into account the non-linear relationships that exist between media investment and its impact on the company’s sales.  \nOn the other hand, in addition to improving accuracy, in the marketing sector it is crucial to maintain full interpretability of what the model is doing. To do this, we have used a combination of two ML models: Generalized Additive Models (GAM) and Gradient Boosting Regressor (GBR) , to which we implemented greater interpretability with the use of SHAP values.  \nIn summary, this thesis develops from the first part of automated data visualization and transformation, going through the creation of linear regression models, which serve as baseline models, in order to compare their performance with the ML models (GAM and GBR) . After the creation of these models, the investment in each marketing channel is optimized to obtain the maximum sales, and this is shown in informative plots so that the company's marketing professionals have a complete overview.  \nKeywords: Marketing Mix Modelling ( MMM) , media variables, baseline variables, adstock transformation, saturation transformation, response curves, regression models, Generalized Additive Models (GAM) , Gradient Boosting Regressor models (GBR) , SHAP values.  \n4  \nDEDICATION  \nFirstly, I would like to thank my tutor Francisco Javier Nogales for helping me during the whole elaboration of this project. Especially for the quick answers and the advice to improve.  \nI would also like to thank my parents for their constant support, which has been key to keep up the hard work during the most difficult days.  \nFinally, I would also like to thank my classmates, who have helped me to make these years some of the best in my life, supporting each other when things didn't go as expected.  \n6  \nCONTENTS  \n1. INTRODUCTION ...………………………………………………………………………………16  \n1.1. Motivation ...……………………………………………………………………………16  \n1.2. Problem statement …………………..……………………………………………….16  \n1.3. Objectives ..……………………………………………………………………………17  \n1.3.1. General objectives ..……………………………………………………….17  \n1.3.2. Specific objectives ..……………………………………………………….18  \n1.4. Structure of the document ...…………………………………………………………18  \n2. CONTEXTUAL FRAMEWORK ...………………………………………………………………20  \n2.1. Regulatory framework ..……………………………………………………………...20  \n2.2 Socio-economic environment ..………………………………………………………20  \n3. STATE OF THE ART ...………………………………………………………………………….22  \n3.1. Previous studies on marketing mix modeling …..………………………………....22  \n3.2. Data science ..………………………………………………………………………...22  \n3.2.1. Data science in marketing ...……………………………………………...23  \n3.3. Machine learning ..……………………………………………………………………24  \n3.3.1. Machine learning applied to marketing ...………………………………..26  \n3.4. Marketing transformations: adstock and saturation ...…………………………….29  \n3.5. Tools and technologies used ..………………………………………………………30  \n3.5.1. Python ...…………………………………………………………………….30  \n3.5.2. Visual studio code and Git Hub ...…………………………………………31  \n4. DATA ORIGIN AND PREPROCESSING ..……………………………………………………32  \n4.1. Data origin ..…………………………………………………………………………...32  \n4.2. Data preprocessing ..…………………………………………………………………33  \n4.2.1. Handling missing values ...………………………………………………..33  \n4.2.2. Adstock and saturation transformations ...………………………………34  \n5. DATA ANALYSIS AND INFORMATION EXTRAC","cbCaib8vgSLMpgkE","https://ap.wps.com/l/cbCaib8vgSLMpgkE","pdf",5558566,88,"English","# INTRODUCTION\n## Motivation\n## Problem statement\n## Objectives\n### General objectives\n### Specific objectives\n## Structure of the document\n# CONTEXTUAL FRAMEWORK\n## Regulatory framework\n## Socio-economic environment\n# STATE OF THE ART\n## Previous studies on marketing mix modeling\n## Data science\n### Data science in marketing\n## Machine learning\n### Machine learning applied to marketing\n## Marketing transformations: adstock and saturation\n## Tools and technologies used\n# DATA ORIGIN AND PREPROCESSING\n## Data origin\n## Data preprocessing\n### Handling missing values\n### Adstock and saturation transformations\n# DATA ANALYSIS AND INFORMATION EXTRACTION\n## Variables information\n## Histograms\n## Boxplots\n## Correlation matrix\n# MODEL DESIGN AND IMPLEMENTATION\n## Statistical approach: Regression models\n### Linear regression\n### Ridge regression\n### Lasso regression\n## Machine learning approach\n### Generalized additive models (GAMs)\n### Gradient boosting regressor (GBR)\n### Final model: GBR- > SHAP- > GAM\n## Investment optimization\n### Optimization techniques (SLSQP)\n### Budget constraints\n# RESULTS OBTAINED\n## Model evaluation\n### Evaluation metrics: MAE, RMSE, R² , MAPE\n### Model selection\n## Visualization of results\n# CONCLUSIONS AND FUTURE WORK\n## Project conclusions\n## Project limitations\n## Proposals for future work\n# PLANNING AND BUDGET\n## Project planning\n## Cost estimation\n# REFERENCES","[{\"question\":\"What is the main objective of the thesis?\",\"answer\":\"To develop a Machine Learning-based MMM approach that improves accuracy over current sector models that assume linear relationships between media variables and sales.\"},{\"question\":\"How does the thesis improve interpretability of the marketing models?\",\"answer\":\"It combines GAM and Gradient Boosting Regressor, then uses SHAP values to provide interpretability for what the model is doing.\"},{\"question\":\"What transformations and modeling baselines are used before final optimization?\",\"answer\":\"The workflow includes automated data visualization and transformation, baseline linear regression models, and marketing transformations such as adstock and saturation before analyzing results and optimizing investments.\"}]","Machine Learning for Marketing Mix Modeling - Improving Accuracy and Interpretability | PDF",222]