Foundations of Data Science

An introduction to basic data science techniques using R or Python

Expected duration

15 to 30 h

Syllabus

Introduction to intermediate programming and data analysis with Python

Scraping data

Introduction to supervised learning:

• High-d regression and classification (e.g. lasso)
• Model selection (predictive, cross validation, information criteria, etc)
• Tree-based models, Ensemble methods (e.g. random forests)
• Additive models and algorithms (e.g. boosting)

Introduction to unsupervised learning

• Singular value decomposition
• PCA, factor models, latent semantic analysis
• Graph-based methods: spectral clustering, diffusion maps

Data Visualization

• Classical data viz methods: correspondence analysis, biplots, multidimensional scaling
• Visualization in Python, visualization in R. Generation of reports/ dashboards.

Prerequisites

• some/moderate experience with data analysis
• some/good background in Statistics/Econometrics
• some experience with computing and programming: we provide an online datacamp tutorial with exercises that have to be completed before course

Credential

The material forms core of Winter School, Summer School Week I and has been taught in various institutions that collaborate with the Barcelona GSE Data Science Center, e.g. CEMFI (Madrid), Universidad de Piura, CAIXABANK, etc