E-mail icon E-mail courses Print icon Print course(s)
 

as of 7/2/2022

BIG DATA: NEW TOOLS FOR ECONOMETRICS
Country - Partner Institution - Programs: Italy - University of Bologna - 'Univ. of Bologna'
UC Course SubjectEconomics
Number & Suffix: 172
Full UC Title: BIG DATA: NEW TOOLS FOR ECONOMETRICS 
Transcript Title: BIG DATA:ECONMETRCS 
UC QTR Units - Division: 6.0 - Upper Division 
Course Description: This is a graduate level course that is part of the Laurea Magistrale program. The course is intended for advanced level students only. Enrollment is by consent of the instructor. Students who complete special term project on a pre-approved topic are awarded 1 extra unit. Maximum units for this course are 6. The course focuses on the main tools used by economists and statisticians in machine learning and statistical learning to analyze large/huge data sets coming from several domains. The course highlights how to apply key aspects of machine and statistical learning, such as out-of-sample cross-validation, regularization, and scalability. Special attention is placed on the concepts of supervised and unsupervised learning, classification, regression, and clustering analysis as well as the detection of association rules. The course also focuses on the main learning tools such as lasso and ridge regression, regression trees, boosting, bagging and random forests, principal components, mixture models and the k-means algorithm. The course places emphasis on the application of the techniques discussed using dedicated open-source software packages on training datasets. Course topics: introduction and overview of statistical learning; linear regression as a prediction tool; binary and multinomial classification: logistic regression, linear discriminant analysis and k-nearest neighbors; resampling methods: cross-validation and the bootstrap; linear model selection and regularization: ridge regression, the lasso, and principal components; moving beyond linearity: regression splines, smoothing splines and general additive models; tree-based methods: CART, bagging, boosting, and random forests; support vector machines and neural networks; unsupervised learning: hierarchical and k-means clustering. The relevant theory will be applied to each topic and subsequently the analysis will move to its empirical application in the R language. Special emphasis is placed on the economic interpretation of the results. The course focuses in several empirical analyses and replicates the results of a few case studies using the statistical software R and several of its packages. 
Language of Instruction: English
 
Partner Title: BIG DATA: NEW TOOLS FOR ECONOMETRICS (LM) 
Partner University Department: Economics 
Partner University Course Number: 81618