Types of data analysis- 4 types

 Data analysis is a crucial part of extracting valuable insights from raw data. Here are four primary types of data analysis:

1. Descriptive Analysis

Purpose: To describe or summarize the characteristics of a dataset.

Features:

- Provides insights into what has happened.

- Summarizes historical data to identify patterns.

- Utilizes measures like mean, median, mode, variance, and standard deviation.

Tools:

- Excel: For basic descriptive statistics and visualizations.

- Tableau/Power BI: For creating interactive dashboards and summaries.

- Python (pandas, matplotlib): For detailed statistical summaries and visualizations.

- R (dplyr, ggplot2): For advanced statistical analysis and visualizations.

Example:

- Analyzing sales data to understand average sales per month.

 2. Diagnostic Analysis

Purpose: To identify the causes of past events and behaviors.

Features:

- Goes deeper than descriptive analysis to find reasons behind trends.

- Uses techniques like drill-down, data discovery, and correlations.

- Aims to answer "Why did it happen?"

Tools:

- SQL: For querying and exploring relational databases.

- Python (pandas, scipy): For performing correlation analysis and hypothesis testing.

- R (stats, caret): For advanced statistical tests and diagnostics.

- Looker/Qlik Sense: For interactive data exploration and diagnostics.

Example:

- Investigating why there was a sudden increase in customer complaints in a particular month.

 3. Predictive Analysis

Purpose: To forecast future events based on historical data.

Features:

- Uses statistical models and machine learning algorithms.

- Identifies patterns and relationships to predict future outcomes.

- Aims to answer "What could happen?"

Tools:

- Python (scikit-learn, TensorFlow): For building and deploying predictive models.

- R (forecast, randomForest): For statistical forecasting and predictive modeling.

- SAS: For comprehensive predictive analytics.

- RapidMiner: For building predictive workflows without coding.

Example:

- Predicting future sales based on past sales data and market trends.

 4. Prescriptive Analysis

Purpose: To recommend actions based on predictive analysis.

Features:

- Suggests decision options to take advantage of future predictions.

- Uses optimization and simulation algorithms.

- Aims to answer "What should we do?"

Tools:

- IBM Decision Optimization: For prescriptive analytics and decision support.

- Gurobi Optimization: For mathematical optimization.

- MATLAB: For optimization and simulation tasks.

- Python (PuLP, pyomo): For creating optimization models.

Example:

- Recommending the optimal inventory levels based on predicted demand and supply chain constraints.

Summary

Descriptive Analysis: Focuses on "What happened?" by summarizing past data.

Diagnostic Analysis: Delves into "Why did it happen?" to understand causes.

Predictive Analysis: Forecasts "What could happen?" using historical data.

Prescriptive Analysis: Advises "What should we do?" based on predictions and simulations.

By understanding and utilizing these types of data analysis, organizations can make more informed decisions, improve operational efficiency, and gain a competitive edge.