1- Terminology and Notation Flashcards
Data definition
- Factual information used as a basis for reasoning, discussion, or/and calculation.
- Numerical in nature
Variable definition
A category of a data set.
Element definition
The entities on which data is collected.
Observation
A set of measurements obtained by a particular element.
Cross sectional data
Statistical analysis which provides information on the characteristics of, and statistical relationships between, individual units of study at a specified moment in time (the moment of data collection).
Cross sectional analysis
Cross-sectional analysis looks at data collected at a single point in time, rather than over a period of time.
Time series data
Data collected over several time periods
Panel data (balanced or unbalanced)
Provides information on individuals both across individuals and over time.
Data issues
- Missing observations (BUT: are encountered in real world, could drop countries from analysis, could empty data- could estimate)
- Outliers
- Unbalanced data- when there is empty data
Assessment of Time Series Data
Good when you are interested in analysing long periods of time for a small number of units. You would conduct the same analysis for say five units and then compare the findings. Not practical to repeatedly conduct the analysis for more than five units.
Assessment of Cross-sectional data
Only provides a snapshot of relationships between variables for one period of time but across a large number of units. Sometimes only cross-sectional data is available e.g. a one-off survey of 10,000 people. Although this would be cross-sectional data and as a result the data set has no time dimension this would be an excellent data set because N is so large.
Assessment of Unbalanced and Balanced Panel Data
Panel data is the best(???) of both worlds as it has a cross-sectional dimension as N is large and also a time dimension because T>1.
