Overview

Description

CALENDAR DESCRIPTION:

Explores how to recognize and learn from patterns in data using modern statistical software for the purpose of economic analysis. Introduces students to techniques for managing, visualizing, and analyzing data to answer real-world economic questions. Students with credit for POL 390, STAT 341, or STAT 452 may not take this course for further credit. Students with credit for ECON 387 under the title "Applied Data Analysis" may not take this course for further credit.

COURSE DETAILS:

Description:  How can we recognize and learn from patterns in data to add value to a firm or to design more effective economic and social policy? During this course, we will learn to use modern statistical software (R) to manage, visualize, and analyze data to answer real-world economic questions. While we will use econometric methods in our programming, the focus of this course is not on the underlying statistical foundations of econometric methods. Those foundations are covered in, among others, BUEC 333 and ECON 435. Assignments, readings, etc. will be posted on SFU Canvas.

Upon satisfactory completion of the course, students will be able to

• Claim proficiency in R, a software environment for statistical computing and graphics (https://www.r-project.org/).
• Perform exploratory data analysis, including data visualization, using complex data downloaded from various sources.
• Analyse data using simple econometric/statistical models to provide answers to real-world economic questions.
• Prepare clear and informative reports that effectively communicate analysis of patterns in social and economic data.

 

Grading:  

Your grade will be based on weekly assignments (70%); and in-person quizzes to ensure the development of basic programming skills (30%). The required in-person quizzes will primarily function as a check to ensure that students are completing the weekly assignments themselves and will be graded pass/fail. Students must attain at least 50% of the total quiz points to demonstrate sufficient learning to pass the course.  Weekly assignments are to be completed using R, in RStudio, using RMarkdown and are required to be independent work. You will hand in your source code, output, and written answers in .Rmd and .html (this will be described in detail through a demonstration during the first lecture). Assignments will require clean and clear communication and a quarter of the score will be for effective communication. Late assignments are not accepted (0 points). For grading, we may randomly select one or two questions from the assignment. Your performance on that question determines your score for that assignment.

Topics:

Weeks 1-8: Data wrangling, data visualization / exploratory data analysis, programming skills

This part of the course will introduce students to using R through RStudio and RMarkdown and will focus on (1) obtaining real-world economic data from various sources; (2) loading, combining and “tidying” the data; (3) exploratory data analysis through data visualization; and (3) programming using iteration, maps, pipes, and functions in R.

Weeks 9-12: Economic analysis of data

The final part of the course will focus on using basic econometric methods to analyse data and will discuss prediction. Students will be introduced to modern machine learning techniques used by data scientists and will apply these techniques to economic data (model selection, LASSO, regression trees, random forests).

Grading

Materials

REQUIRED READING: