Coding for Applied Economists

Introduction to Stata

Notes

Remind students to install Stata

Have students open Github Desktop

Have students open Stata.

Stata

  • Programming language
  • IDE - Integrated Development Environment
  • A statistics and data analysis environment
  • Now can run simulations, do ML, produce markdown files, interact with Python
  • R is what runs all of the code we will write this semester
  • Separate from RStudio

Stata Graphic User Interface (GUI)

  • Three components for interfacing with Stata
    1. Command line - contains several windows
      • Command window
      • Results window
      • History window
      • Variable window
      • Results window
    2. Project Manager - contains several windows
      • Project window
      • Text Editor window
      • Navigator window
      • Properties window
    3. Data Editor/Browser
    4. Viewer - how one reads help files and logs
    5. Graph Editor
  • Command window is where Stata is actually running
    • Can work in here “interactively”
    • Run a single command and see the result
    • This is also where Stata will run code written in the text editor
  • Text editor (.do-files)
    • Where we write code we want to keep and potentially reuse later
    • Creates a plain text file that stores the code we’ve written
    • This is where you will write all your code and is what you will submit for assignments
  • Can also run commands using Stata’s drop-down menu
    • Never do this to actually execute commands!
    • can be useful to learn syntax of new commands or infrequently used commands

Basic expressions

  • Write code directly in the text editor
    display 2+2
    display 2 * ttail(20, 2.1)
  • This is called an expression

  • A set of commands that returns a value

  • Run line

  • Run selection

  • Save as stata-intro.do
  • Need to add file or directory to see it in the Project window
  • We can also use this tab to create, delete, and rename files & folders

Loading data

  • Load sample data file
    sysuse auto.dta, clear
    describe
  • Stata files end in .dta
  • Stata can input data in almost any format

  • You can look directly at the data using Stata’s Data Editor/Browser

  • Look at the data
    sort price
  • Take a look at the data again

Variables

  • Data sets will already contain variables

  • A variable is a name that has a value associated with it

  • We can manipulate an existing variable
    sum weight
    replace weight = weight * 2.205
    sum weight
  • We can also create new variables
    gen kmpl = mpg / 2.3520
    sum mpg kmpl
  • Often more useful than looking at the data in the data frame is visualizing the data
  • Stata has numerous different graphics commands
  • However, most start with graph twoway
    graph twoway (scatter price mpg) (lfit price mpg)

Basic Stata syntax

    prefix: command [varlist] [= exp] [if exp] [in range] [weight] [, options]
  • prefix: A prefix to a Stata command that modifies how or on what the command is carried out, such as by or quietly.
  • command: The specific Stata command you want to run, such as summarize, list, or graph. Most commands can be shortened (e.g., sum for summarize).
  • [varlist]: A list of one or more variables to be used by the command.
  • [=exp]: An optional expression to create a new variable.
  • [if exp]: A conditional statement that restricts the command to only certain observations. The expression is evaluated as true or false. For example, if rep78 > 3.
  • [in range]: A qualifier that selects a specific range of observations, such as in 1/100 for the first 100 observations.
  • [weight]: Specifies a weighting variable.
  • [, options]: A comma followed by options that modify the command’s behavior. Options are separated by spaces. For example, summarize, detail provides more detailed statistics.

Comments

  • Remember what code is doing
  • For humans, not computers
  • Use the *
* Convert imperial to metric

Assignments format

  • SEPARATE FILES FOR IN-CLASS CODING AND HOMEWORK
  • Comment before each problem and each sub-problem
**# Problem 1

**## 1.1 - convert length to metric

    replace     length = length * 2.54

**## 1.2 - compare mpg across car type
    sum         mpg, by(foreign)

**# Problem 2

* regress price on mpg and repair record
    reg         price mpg rep78

Create assignment script, put in new folder

  • Now we’re going to work on some exercises to get a feel for this
  • In class we will often only do part of an exercise and save the rest for later

Do Exercise 1.1-1.3 - Basic Expressions

Do Exercise 2 - Basic Variables

Summary

  • Stata is both a programming language and an Integrated Development Environment (IDE).
  • Stata allows us to manage data, run statistical analyses, and produce deliverables.
  • The GUI includes specialized windows for commands, results, variables, and history.