ECON1203 Lecture Notes - Lecture 1: Time Series, Data Set, Long Tail
Week 1 – Descriptive Statistics
Types of data
• Variable – characteristic of a population or of a sample from a population
Can observe values or observations of a variable
Data set contains observations on variables
• Variables may be:
Discrete or continuous
Quantitative (numerical) or qualitative (categorical; nominal, ordinal)
(a) Ordinal qualitative data feature a natural order
• To apply statistical analyses directly to qualitative data, we must convert it somehow
to quantitative data
Types of observations
• Time series data consist of measurements of the same concept at different points in
time
• Cross sectional data consist of measurements of one or more concepts at a single
point in time (many variables)
Frequency distributions – categories need to be mutually exclusive and exhaustive
Histograms – data is ordinal, create categories or classes by defining lower/upper class
limits (mutually exclusive and exhaustive)
• Bins need to be of equal width, may be open ended at top or bottom
Describing histograms
• Symmetry – left side is the same as right (i.e. bell-shaped curve)
• Skewness – feature of an asymmetric histogram
Long tail to right: positively skewed
Long tail to left: negatively skewed
May be associated with outliers
• Number of modal classes/bins
Modal class is the class with highest frequency
Histograms may be unimodal or multimodal
Bivariate relationships
• Contingency table – captures relationship between two qualitative variables
• Scatterplots – captures relationship between two quantitative variables
If one variable is time, we get a time series plot
No slope = no relationship
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Types of data: variable characteristic of a population or of a sample from a population. Can observe values or observations of a variable. Data set contains observations on variables: variables may be: Quantitative (numerical) or qualitative (categorical; nominal, ordinal) (a) ordinal qualitative data feature a natural order: to apply statistical analyses directly to qualitative data, we must convert it somehow to quantitative data. Types of observations: time series data consist of measurements of the same concept at different points in time, cross sectional data consist of measurements of one or more concepts at a single point in time (many variables) Frequency distributions categories need to be mutually exclusive and exhaustive. Histograms data is ordinal, create categories or classes by defining lower/upper class limits (mutually exclusive and exhaustive: bins need to be of equal width, may be open ended at top or bottom.