Scribd
Upload
15 views13 pages

Lecture 3 Economics Dissertation - 2025

The lecture covers types of data (cross-sectional, time series, and panel) and variables (numeric and categorical) relevant for economics dissertations. It introduces STATA software for data analysis, including installation, data compilation, and common econometric issues. Practical demonstrations and resources for using STATA are also provided to assist students in their dissertation work.

Uploaded by

hbaecon
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views13 pages

Lecture 3 Economics Dissertation - 2025

The lecture covers types of data (cross-sectional, time series, and panel) and variables (numeric and categorical) relevant for economics dissertations. It introduces STATA software for data analysis, including installation, data compilation, and common econometric issues. Practical demonstrations and resources for using STATA are also provided to assist students in their dissertation work.

Uploaded by

hbaecon
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Economics Dissertation

[ECO00045H]

Lecture 3: Data, methods and using STATA

Dr Nahid Farnaz
[email protected]

1
Lecture 3 - outline

To discuss different types of data and variables

To demonstrate the use and application of Stata

To outline some common econometric issues

To discuss the marking breakdown for the final dissertation

2
Types of data

• Cross sectional data: observing units (individuals, households, firms,


countries, etc) at a particular time (e.g. 2020). This means data collected at
a single point in time across multiple entities (for e.g. household income
survey in 2020)

• Time series data: observing a phenomenon (for e.g. macro indicator) for a
country (or countries) over a period of time This means that data collected
over time for a single entity (e.g. UK inflation rate from 2000-2023)

• Panel data (longitudinal data): pooling of observations on a cross-sectional


of households, firms, countries, etc over several time periods. This
combines cross-sectional and time series aspects – tracks multiple entities
over time (e.g. GDP and inflation rates for OECD countries from 2000-2023)

3
Types of variables
Numeric variables: can be measured numerically (e.g. income, height,
temperature)
• Discrete: Values can only take on specific, distinct values (usually whole numbers), often
representing counts. Example: number of students in a class, number of cars in a parking lot,
number of siblings.
• Continuous: Values can take on any value within a given range (e.g., height, weight, temperature).

Categorical variables: represent groups or categories that cannot be measured


numerically (e.g. gender, country, industry type)
• Nominal: Categories have no inherent order or ranking (e.g., colours, types of vehicles, gender).
Example: Hair colour (brown, blonde, black)
• Ordinal: Categories have a meaningful order or ranking, but the intervals between them may not
be equal (e.g., satisfaction levels, education levels, rating scale). Example: Survey responses on a
Likert scale (strongly agree, agree, neutral, disagree, strongly disagree).

4
Slido poll #

Which type of data are you working with for your dissertation?

a) Cross sectional

b) Time series

c) Panel

d) Not sure yet!

5
Getting started with STATA

6
Installing STATA

● Stata is on all university IT managed computers on campus.


● University has a license to install on your personal laptop if you are
in one of the departments in the Social Science Faculty
● To install, go to the IT software Stata pages

7
Stata interface
● Open STATA
● Results window (centre)
● Review window (left)
● Menus (top)
● Command window (bottom)

● The command is repeated as well as the answer in the results window


● Separate windows will appear for viewing data, plotting graphs, and
writing commands

8
Using STATA: useful resources

• Stata Cheat Sheets

• LSE STATA Tutorials (videos)

• Statology Stata Guides

9
Data compilation and cleaning
Importing and cleaning data:
• Loading and browsing data (opening a data file and browsing its contents), changing
variable names
• Check for missing data: identify missing values and handle missing data (misstable
summarize, drop if, replace)
• Merge datasets: merging two datasets together, or append (adding new data)

Basic data transformations:


• Transforming a variable (for e.g. generating lagged values or log of a variable)
• Generate new variables: gen, replace, label variable
• Creating dummy variables

Top tip: use the “help” command for more information on a STATA code (simply type
“help” followed by the code in the command window on STATA)
10
Using STATA: do Files
● Using do-files: Type your commands and comments in a .do file.
● They are a record of what you have done (There would be no record if you use
the menus instead of commands.)
● To open a .do file either:
○ Click pencil and pad icon for a new file or
○ File > Open and navigate to a previously saved .do file
● Remember to save your .do file - this is not done automatically

11
Stata demonstration (practical session)

• Cross sectional data

• Time series data

• Panel data

12
Common data problems

• Heteroscedasticity

• Multicollinearity

• Autocorrelation

• Omitted Variable bias

• Endogeneity

13

You might also like