EDA objective:
We will look for patterns, differences, and other features that address the questions we are interested in. At the same time we will check for inconsistencies and identify limitations.
Types of Data
- Nominal : Labels
- Ordinal: Qualitative or Ordered values. Limited functions of stats. No meaning of "true zero:
- Interval Scale: Values with an order and distances. Discrete Data like year.
- Ratio Scale: Real Numbers.Meaning of true zero.
Type of Data Presentation:
There are three modes of presentation of data i.e. textual presentation, tabular presentation, and diagrammatic presentation.
Diagrammatic presentation:
Histogram:
- Datatype: Discrete/Continuous Data
- Used for probability density function (frequency density)
- No space between bins to indicate the continuous nature of values.
- Indicates the distribution of the discrete data.
- The value is proportional to the area of the bar.
- Uni-variate Analysis
Bar Chart:
- Datatype: Categorical Data
- Relates between two variables.
- Multi-variate analysis
Scatter (Multi-Variate) :
- Datatype: Continuous Vs Continous
- Relation (liner or non-linear) between pair of variables.
- visual representation of correlation
Boxplot:
- Datatype: Continuous
- The points outside of whisker are designated as outliers.
- SIQR (Semi-Interquartile range).
- Most of the data is expected in the range of median 土 3SIQR
What is its relation to the Normal Distribution ? - Uni-variate Analysis
Side-by-Side Boxplot (Multi-Variate)
- Data type: Continuous vs Categorical
- Box plots for each level of categorical variables (Ozone vs Month). No statistical relevance
Tabular presentation
Contingency Tables (Multi-variate analysis):
- Data type: Categorical Data Vs Categorical Data
- Relation between Categorical Variables (count and count%)
- Contingency tables can be analysed for association between rows an columns using the chi-squared test
Problems
- The randomness of the data presents problem
- Missing values treatment
Cheat Sheet (Linear Data):
Data Type | One Variable | Multi Variable | Visual |
| |||
Continuous Vs Categorical | No |
| |
| |||
Categorical Vs |
|