Archives

Types of Analytics

Jan2

Data is a vital asset for any organisation. Data can be raw, unstructured and unorganised. To extract meaningful insights and information from data we need to analyse it. There are different types of analytics.

Data Analytics is the process of cleaning, reviewing, purifying, modelling, transforming

Descriptive analysis

This analysis looks at what happened in the past. It does not explain why it happened , or any cause effect relation or any future analysis. It allows you to condense large information into smaller bits. It helps us to understand information on an aggregate level. It is used to summarize and describe the data’s main features. Descriptive analysis helps an organisation learn from past behaviour and use this data for future analysis.

E.g :

We have a software. We may use descriptive analysis to summarize and describe data collected from people who use the software.

Diagnostic analysis

This analysis looks at why the data happened? It helps to identify and respond to anomalies in our data. It is a root cause analysis where we find out the reasons for the occurrence of a particular event. We can identify patterns, isolate patterns of data. This analysis is a deep dive into data and find out why the data metrics a particular way.

E.g :

If we have a 30% increase in our sales in a particular region. We need to find out why this happened. We can further breakdown sales into product categories, sub categories to do a further deep dive. We can identify where sales increase .

Predictive analysis

This analysis looks at what is likely to happen in future?

We look at past data patterns , we use predictive models to estimate the likelihood of an event. Statistical modelling, algorithm, machine learning are used for predictive modelling. Relation between set of variables is used to make predictions. Data gathered during descriptive and diagnostic phase is used for prescriptive analysis.

E.g :

  • We can use relation between promotional schemes and sales to predict if a particular scheme will lead to increase in sales.
  • We can predict on a festival weekend what will the orders placed in a restaurant.
  • We can analyse our competitors behaviour to predict future behaviour

Prescriptive analysis

This analysis looks at what action should be taken?

We can use machine learning, statistical modelling and algorithm to make decision patterns that a company might take. We can use a combination of conditions, methods to measure the impact of a decision. We can find out the best route of decision.

E.g

What should be course of action to increase sales basis past pattern of data.

Data and its Types

Jan1

 

Data is a collection of raw or organized information which can be in the form of numbers, characters, images, video, audio, bytes, documents and several other forms. They can be facts, figures, measurements, graphs, raw unprocessed information etc.

TYPES OF DATA

QUALITATIVE DATA

Qualitative data describes the data that fits into the categories. These are non-numerical in nature that describes qualities, characteristics, or concepts. Examples include a persons gender, town.

Sometimes categorical data can be numerical values (quantitative value), but these values do not have a mathematical sense.

They are further classified into Nominal Data and Ordinal Data.

Nominal Data

Nominal data is data that consists of categories or names that cannot be ordered or ranked. They are used to categorize the observations into groups, and these groups are not comparable.

E.g :

  • Gender Groups : Male, Female, Undisclosed
  • Ethnic Groups : White , Asian, Hispanic, White, Alaskan Native
  • Hair Colour : Black , Brown, Blonde
  • Nationality : Indian, American, French
  • Transport : Bus, Train, Taxi
  • Brands : Brand A, Brand B, Brand C

Ordinal Data

Ordinal data is data that consists of categories or names that can be ordered or ranked. Examples of ordinal data include education level. Ordinal data can help to compare one item with another by ranking or ordering

E.g :

  • Grade Level : Grade I , Grade II, Grade III, Grade IV
  • Education Level : SSC, HSC, Graduation, Masters, Phd
  • Seniority Level : Junior, Mid, Senior

QUANTITATIVE DATA

Qualitative data describes the data represents the numerical value of the data.

E.g :

  • Height , Weight , Room Temperature, Test Scores, Time
  • Number of students in a class
  • Number of Products in a Shop

They are classified into Discrete Data and Continuous Data.

 Discrete Data

The discrete data is data that is distinct and separate. It is data that have a particular value rather than a range.They have countable values and are finite. Discrete contain values that fall under integers or whole numbers. Discrete Data is Countable. There are distinct or different values in discrete data.

E.g :

  • Number of employees in a company

Continuous Data

They contain values that represent the data in a continuous range. The variable in the data set can have any value between the range of the data set. Continuous Data is Measurable. Every value within a range is included in continuous data. The data that lies between the highest and the lowest value are called the continuous data. The range of the continuous data is the difference between the highest and the lowest data.

E.g :

  • Temperature Range
  • Salary Range
  • Time
  • Length
  • Mass
  • Capacity