Predictive analytics describe the use of statistics and modeling to determine future performance based on current and historical data. Predictive analytics look at patterns in data to determine is event can be occur in future or what problem can be occur.
Different techniques in predictive analytics
- Statistical Foundation
- Inferential Statistics
Statistical is used to extract meaningful information from data by applying mathematical computation on data. Math and stats are building block or core of machine learning models / algorithms.
Types of Statistical
Measure of central Tendency
- Mean: Average of data
- Median: Middle number of data (1,2,(3+4/2),5,6,)
- Nominal (Order not matter)
- Ordinal (Order of data matter)
- Discreate (Data that can not be in decimal, eg: 4,5,6,7)
- Continues (Data that can be in decimal or float. eg: 1.2, 3.5)
Example of data types
Nominal Data: Gender, Colors, Pass or Fail
Ordinal Data: Student Roll No, Pass Division.
Discreate Data: Number of family members, total number of pass student in calss.
Continues Data: How much milk in bucket,