Exploratory Data Analysis

&

Data Cleaning

Agenda - I

  • Univariate Analysis
  • Bivariate Analysis
  • Multivariate Analysis
  • Outlier Detection

Uni-variate Analysis

Description, summarization or inspection of singleton variable 

Uni-variate Analysis

Description, summarization or inspection of singleton variable 

Tools for Uni-variate Analysis

Continuous Variable

Discrete Variable

  • Mean
  • Median
  • Range
  • Variance
  • Standard Deviation
  • Z-Test
  • T-Test
  • Chi-Square Test
  • Missing Values
  • Frequency Distribution Tables
  • Mode
  • Kolmogrov - Smirnov Test
  • T - Test for proportions
  • Missing Values

Summary Statistics

Z-Test Vs T-Test

Chi Square Test

Frequency Distribution

Mode

KS Test

Bi-variate Analysis

- Understand relationship between two variables.

 

- Can be performed for any combination of categorical & continuous variables.

Bi-variate Analysis

Continuous - Continuous

 

Bi-variate Analysis

Continuous - Categorical

Bi-variate Analysis

Categorical - Categorical

 

Outlier Detection

Exploratory Data Analysis / Data Cleaning

By Rahul Bajaj

Exploratory Data Analysis / Data Cleaning

  • 116