ACM: comprendre enfin l'analyse des correspondances multiples

Does Multiple Correspondence Analysis (MCA) intimidate you? Are you unsure how to interpret a scatter plot? In this video, we'll transform statistics into a field investigation! [Video Content] Using the "Poison" dataset, we'll identify the food responsible for a case of food poisoning without performing a single regression. I'll explain step by step how to: Read a Scree Plot in MCA (and why 40% inertia is already excellent!). Interpret the proximity between categories (the link between Mayo Clinic and Illness). Use additional variables (Qualitative and Quantitative) to enrich your analysis without skewing the results. Include additional individuals to compare atypical profiles. [Chapters] 0:00 - MCA: The Investigation Begins 0:30 - Why Isn't MCA a Principal Component Analysis (PCA)? 1:00 - Pitfalls in ACM 5:30 - The Modalities Plan: Who is close to whom? 8:00 - The Scree Plot: The Trap of Low Percentages 9:00 - Additional Variables (The Culprit Unmasked) 13:45 - Additional Individuals: Analyzing Specific Profiles 13:40 - Interpreting Confidence Ellipses 14:00 - Conclusion: What to Write in Your Thesis [Links & Resources] 📥 Download the Poison Dataset: [Link] 💻 My Complete R Code (FactoMineR): [Git/GitHub Link: Leave a Comment] 📚 My Video on PCA: [   • ACP : Comprendre enfin l’Analyse en Compos...  ] #Statistics #MCA #DataAnalysis #FactoMineR #Rstats #Econometrics #Noukhaus #DataAnalysis