Econometric and Statistical modeling

Part 1 – Applied intro to R

a) Data import, export, saving codes, results, graphics

b) Typical operations in data analysis

c) Data preparation (cleaning, imputation)

d) Using packages

e) Reproducible research

Part 2 – Applied statistical modeling

[1] Statistical distributions – types of probability distributions, generating random variables, testing consistency, differences and similarities, usage of probability distributions

[2] Inequality measures – one-dimensional and two-dimensional measures (e.g. Gini, Herrfindahl, entropy, KLD, mutual information)

[3] Monte Carlo simulations – formulating the problem and underlying distributions, aggregation and distributions of results, confidence intervals, sensitivity analysis

[4] Bootstrapping – sampling and replications issues, aggregation of independent and inter-dependent results, confidence intervals, sensitivity analysis, strata sampling

[5] Survey data analysis - discriminant analysis, factor analysis, non-parametric tests, different measurement scales, rand index, mantel test

Part 3 – Applied econometric modelling

[6] Regression and hierarchical linear models - model and variable selection, estimation, testing, forecasting, missing data issues

[7] Panel regression models - model and variable selection, estimation, testing, forecasting, missing data issues, cross-validation

[8] Propensity score matching - model and variable selection, estimation, testing, forecasting, missing data issues

[9] Difference in difference, regression discontinuity – estimation, missing data management, quality of fit

[10] Bootstrapped, jackknife, quantile regression – sampling and replications issues, interpretation of results, forecasting

[1] Damodar, G. (2013). Econometrics by example, The McGraw-Hill/Irwin Series in Economics

[2] James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. New York: Springer.

http://faculty.marshall.usc.edu/gareth-james/ISL/ISLR%20Seventh%20Printing.pdf

[3:n] Papers selected for classes

1) Each student is to review (orally) the assigned paper on selected quantitative topic during the class – extra points are given to the audience for discussion

2) Attendance is obligatory – for each absence students write the paper on omitted topic (Part 2 & 3) (topics will be given before the class)

3) Students are to find a paper which deals with “their” topic and uses the quantitative methods and to replicate (and possibly develop) the study on similar data (collected or generated)

4) Points collection:

a. max. 20 points review & oral presentation of assigned paper,

b. max. 20 points – activity during classes (esp. discussion of others reviews)

c. max. 60 points – own paper as in 3)

d. 0-1 criteria – attendance or papers for absence

5) Grading scale: 0%-50% - 2 (negative), 50%-60% - 3 (sufficient), 60%-70% - 3+ (more than sufficient), 70%-80% - 4 (good), 80%-90% -4+ (more than good), 90%-95% -5 (very good), 95%-100% - 5! (with honours)