TEACHING
TEACHING POSITIONS
Teaching Fellow,Columbia University
Sep 2019  Presentâ€‹

Economics Department, Graduate School of Arts and Science
Econometric Courses
Industrial Organization Courses

School of International and Public Affairsâ€‹â€‹
Energy Economics Courses
Development Economics Courses
NOTES AND CODES
â€‹Econometric and Quantitative Methods Notes
Available Upon Request
Summary: The goal of these notes is to introduce and discuss different statistical methods used to conduct econometric analysis. The first part of these notes explains the concepts of probability theory, statistical modeling and inference in detail. I describe basic concepts as random variables, estimation, hypothesis testing, and confidence intervals. The second part of the notes describe linear models (OLS, Restricted Least Squares, GLS, Instrumental Variables, Panel Data, and Dynamic Panel Data). The third and fourth part discuss nonlinear econometrics methods including but not limited to nonlinear least squares, GMM, quantile regression, MLE and limited dependent variable models. The rest of the notes discuss causal inference and special topics as selection bias and truncated data, density estimation, Bayesian statistics, among others. The notes also cover machine learning methods as Lasso, Ridge, Random Forests and Neural Networks.
â€‹
Describe your image
Describe your image
â€‹Sample R Code (Demand and Production Function)
To access codes, click on the buttons on the right side of the screen
â€‹Summary: The codes for the production function estimate this function using fixed effects, OlleyPakes (OP), LevinsohnPetrin (LP), and AckerbergCavesFrazer (ACF). Standard errors are calculated using the GMM standard errors and bootstrapped. Three codes can be used to estimate the demand function. The first estimates the demand following Bresnahan's vertical model. The second code uses a logit model and the last code implements the BerryLevinsohnPakes (BLP) model. Standard errors are bootstrapped. Data files are provided.