AI/ML & Data Science

# Types of Data in Data Analysis

Broadly data is categorized into three types, namely

• Numerical – Quantitative
• Categorical – Nominal
• Time Series

## Numerical – Quantitative

Data are numbers and are not a function of any other quantity, such as time. Data that holds a proportional value of its representation.

Numerical data can be of two types.

• Discrete
• Continous

Discrete

The data can hold complete values as in integers such as 10, 25, 47, 82 …etc. It is considered as discrete. Let’s say a cricketer, a batsman can score only a whole number of runs such as 23, 65, 89 ..etc. He/She can not score 43.53 runs. Any type of data which can hold integer values is considered discrete. E.g., Whole numbers ( 2, 5, 6, 23, 534, …etc.)

Continous

The data which can hold floating-point values, i.e., continuous values, is termed as continuous data. The batting average of a cricketing batsman can be 43.0 or 37.32 …etc. Such type of data which can hold floating values is termed as continuous. E.g., Any value in a range (2.0, 4.23, 53.244, …etc.), the weight of a person.

## Categorical – Nominal

Data that represents characteristics, for example, the color of a fruit, group of a candidate, the position of a cricket fielder, team, …etc. Categorical data can hold numbers, but they are representational. They don’t have meaning in a mathematical sense. For, e.g., one can not perform mathematical operations on categorical data. They just represent a characteristic. Although there is something called ‘ordinal data.’

Ordinal

Ordinal data is ordered categorical data. The best example of ordinal data grades, e.g., A, B-, B, C, C-, D, F these are representational letters for a group which your score falls in, but they do have order in which we know B is better than C- (c minus) and F is failing.