Course 3

COURSE 3 · ADVANCED · 8 TOPICS · TUTORIALS

Mathematical & Statistical Methods

The mathematics and statistics the models rest on, each implemented in R.

What you’ll be able to do by the end

  • Use eigenvalues, the SVD and graph Laplacians to analyse the matrices that arise in games and networks.
  • Solve linear and dynamic programmes and compute fixed points numerically.
  • Quantify information with entropy, mutual information and KL divergence.
  • Carry out frequentist inference, regression and nonparametric tests.
  • Fit Bayesian models with conjugate priors and MCMC.
  • Estimate causal effects with DAGs, difference-in-differences, instrumental variables, regression discontinuity and synthetic control.
  • Model time series with ARIMA, VAR and GARCH, and analyse random graphs, centrality and communities.

Who should take this course

For readers who want the methods behind the models. Courses 1 and 2 supply the examples but are not required; comfort with R and algebra is.

The eight topics

Topic 1 · tutorials

Linear Algebra & Matrices

Eigenvalues, SVD, positive definite matrices, Perron-Frobenius, graph Laplacians.

Topic 2 · tutorials

Optimization & Numerical Methods

Convex optimisation, LP, dynamic programming, gradient methods, fixed-point.

Topic 3 · tutorials

Information Theory

Entropy, mutual information, KL divergence, channel capacity, rate-distortion.

Topic 4 · tutorials

Statistical Foundations

Probability, inference, hypothesis testing, regression, nonparametrics.

Topic 5 · tutorials

Bayesian Methods

Conjugate priors, MCMC, Stan, hierarchical models, Bayesian computation.

Topic 6 · tutorials

Causal Inference

DAGs, potential outcomes, DiD, IV, RDD, synthetic control.

Topic 7 · tutorials

Time Series & Econometrics

ARIMA, VAR, cointegration, GARCH, structural breaks, panel data.

Topic 8 · tutorials

Network Science

Random graphs, centrality, community detection, epidemics, igraph/tidygraph.

Contents

Every tutorial in this course, topic by topic →

The other courses

Back to top