Scribd
Upload
13 views6 pages

Data Analytics

The document outlines the objectives and methods of data analytics, including descriptive, diagnostic, predictive, and prescriptive analytics. It details the data analysis process and fundamental statistical methods such as distribution, deviation, correlation, regression, and clustering, emphasizing their applications in uncovering insights and making informed decisions. The importance of good data quality is highlighted as essential for effective analysis.

Uploaded by

anviza
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
13 views6 pages

Data Analytics

The document outlines the objectives and methods of data analytics, including descriptive, diagnostic, predictive, and prescriptive analytics. It details the data analysis process and fundamental statistical methods such as distribution, deviation, correlation, regression, and clustering, emphasizing their applications in uncovering insights and making informed decisions. The importance of good data quality is highlighted as essential for effective analysis.

Uploaded by

anviza
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Data Analytics : Analyze data and uncover actionable insights

Objectives of data analytics :

- Explore patterns in data


- Predict future trends
- Make informed decisions
- Strategize with deeper understanding

Descriptive : Discover what happened in the past

Diagnostic : Explain why it happened

Predictive : Predict what will happen in the future

Prescriptive : Advice about actions and decision to be taken

Descriptive Data Analytics (What Happened ?)

- Gather the data


- Aggregate the data
- Find patterns in the data

Diagnostic Data Analytics

- Explain what was uncovered in the descriptive analysis


- Describe any causal relationships discovered

Predictive Data Analytics

- Predict Future events


- Historical data and probability
- Key business decisions

Prescriptive Data Analytics

- Advice regarding actions and decisions to take


- Leverage insights uncovered by predictive analysis

Data Analysis Process

- Define questions
- Collect data
- Clean data
- Analyze data
- Provide insights
BASIC ANALYTICS METHODS

UNDERSTANDING DISTRIBUTION

5 Fundamental methods of statistical analysis


- Increase confidence
- Make comfortable with analysis
- Shared vocabulary

Distribution : how often something happens (Frequency of occurrences), probability distributions

3 Measures of central tendency :

1. Mean – numerical average


2. Median – Midpoint on the scale
3. Mode – highest number of occurrences
Distribution analysis is very useful, provide insight into the frequency and patterns of business
events

All of this methods rely on good and sufficient data

Statistical method Benefits


Distribution Distribution tells you how often things happen.
It measures the frequency of occurrences
using a scale. Using distribution, you can
measure the mean, median, and mode of a
range of values, which enables you to discern
patterns in events.
Deviation Deviation measures how far things are from
the average in a range of values. That means
that deviation analysis helps provides insight
into how much events differ from the norm.
Correlation Correlation measures how two variables track
together using a scale from -1 to +1. A strong
(+1) correlation means there’s a strong
relationship between the two variables.
Correlation can indicate what impact
changing one variable will have on another
variable.
Regression Regression analysis quantifies how much
influence on average a given variable has on
a specific business event or outcome. It can
investigate causal analysis between
dependent and independent variables.

Regression analysis can be used in


forecasting and predictions based on trends.
It also helps identify performance drivers.
Clustering Clustering groups objects so that they are
similar to other objects in their clusters and
dissimilar to the objects in other clusters.

Understanding clustering is beneficial


because it provides insight into how business
objects and events form natural groups that
can receive individualized treatment.

Understanding Deviation
Deviation : How far things are from the average

Deviation also presented by percentile distribution with test score.

Standard deviation of 1 includes 68% of occurrences

Standard deviation of 2 includes 95% of occurrences

Standard deviation of 3 includes 99.7% of occurrences

Standard deviation curves are bell-shaped

Variance is a measure of volatility and risk

Identify the ways in which deviation may be represented is Range


and percentile distribution

Understanding Correlation
Relationship between 2 or more variable (How two variables track together= simple covariance)

Positive covariance = 2 variable track together (As One variable goes up, the other variable goes up)

Negative covariance = Variables track together, but in opposite directions (As One variable goes
down, the other variable goes up)

Zero covariance = The two variables are uncorrelated, independent


Standard = Comparable

Don’t take correlation to mean causation

Meaning of correlation depends on context

Data points are clustered loosely together on a slope = Weak


Data points are clustered closely together on a slope = Strong
Data points form an upward slope on the graph = Positive
Data points are widely spread out on the graph = Uncorrelated (In other
words, the variables are independent of each other.)

Data points form a downward slope on the graph = Negative

Understanding regression
Regression analysis is Quantifies how much influence

Regression is similar to correlation

Dependent variable (Sumbu Y) is influence dependent one ex. sales

Independent variable (sumbu X) is Price selection, promotion, depending on product, customer


satisfaction

Linier regression = effect of single independent variable

Uses two or more independent variables to quantify and predict an


outcome = Multiple linier regresion
Works with data that describes a curve rather than a straight line = Non-
Linier regression

Understanding Clustering
Many application:

- Marketing
- Products
- Customer service
- Training
- Financial services

The clusters are algorithmically discovered. Clustering is pattern recognition and data mining
Goal of clusters is to minimize intra-cluster distances so that objects in a cluster are as similar as
possible and maximize inter-cluster distances so that clusters are as distinct as possible

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy