NAME: RIYA GHOSH
ROLL NO: 23460421001
COURSE: BSC IN STATISTICS
SEMESTER: SIXTH
PAPER NAME: MULTIVARIATE ANALYSIS AND
NONPARAMETRIC METHODS
PAPER CODE: BSTAT 602
1. What is multivariate analysis?
In data analytics, we look at different variables (or factors) and how
they might impact certain situations or outcomes.
For example, in marketing, you might look at how the variable “money
spent on advertising” impacts the variable “number of sales.” In the
healthcare sector, you might want to explore whether there’s a
correlation between “weekly hours of exercise” and “cholesterol
level.” This helps us to understand why certain outcomes occur,
which in turn allows us to make informed predictions and decisions
for the future.
There are three categories of analysis to be aware of:
• Univariate analysis, which looks at just one variable
• Bivariate analysis, which analyzes two variables
• Multivariate analysis, which looks at more than two variables
As you can see, multivariate analysis encompasses all statistical
techniques that are used to analyze more than two variables at
once. The aim is to find patterns and correlations between several
variables simultaneously—allowing for a much deeper, more
complex understanding of a given scenario than you’ll get with
bivariate analysis.
2. How is it different from bivariate analysis?
POINT OF BIVARIATE MULTIVARIATE ANALYSIS
DIFFERENCE ANALYSIS
1. Number of Involves the Expands the scope to
Variables examination of analyze relationships
relationships among three or more
between two variables simultaneously.
variables.
2. Complexity of Focuses on simple Allows for the exploration
Relationships: relationships of complex interactions
between two and dependencies among
variables. multiple variables.
3. Depth of Provides a basic Enables a more
Understanding: understanding of comprehensive
the association understanding by
between two considering the joint
variables. influence of multiple
variables on an outcome.
4. Statistical Commonly Utilizes advanced
Techniques involves simpler statistical methods such
statistical as multiple regression,
techniques like factor analysis, or cluster
correlation and analysis to handle multiple
regression for two variables.
variables.
5. Data Interpretation Requires careful
Interpretation tends to be more interpretation considering
straightforward the interplay of multiple
due to the focus on variables, demanding a
a pair of variables. nuanced understanding of
their combined effects.
3. Give examples of multivariate analysis.
Example 1: Let’s imagine you’re interested in the relationship
between a person’s social media habits and their self-esteem. You
could carry out a bivariate analysis, comparing the following two
variables:
a) How many hours a day a person spends on Instagram
b) Their self-esteem score (measured using a self-esteem scale)
You may or may not find a relationship between the two variables;
however, you know that, in reality, self-esteem is a complex
concept. It’s likely impacted by many different factors—not just
how many hours a person spends on Instagram. You might also
want to consider factors such as age, employment status, how
often a person exercises, and relationship status (for example). In
order to deduce the extent to which each of these variables
correlates with self-esteem, and with each other, you’d need to run
a multivariate analysis.
So we know that multivariate analysis is used when you want to
explore more than two variables at once. Now let’s consider some
of the different techniques you might use to do this.
Example 2: Imagine you work for a retail company that sells a wide
range of products.
The company has gathered data on various customer attributes
such as age, income, spending habits in different product
categories (e.g., electronics, clothing, groceries), and their
frequency of visits to the store.
The goal of the company is to understand its customer base better
in order to tailor marketing strategies and product offerings more
effectively. They want to identify distinct customer segments based
on these attributes.
Here’s how multivariate analysis can be applied:
• Data collection: Conduct an online survey to collect data on
customer attributes: age, income, spending habits, visit
frequency, etc.
• Applying multivariate analysis: The analysis will identify new
“principal components” that capture the most significant
patterns of variation in the data.
• Customer segmentation: Using the scores on the retained
principal components for each customer, you can now perform
clustering techniques to group customers into market
segments based on their similarities in terms of the principal
components.
Tailor marketing strategies for each customer segment: For
example, if one segment is characterized by high spending on
electronics, the company might target them with electronics
promotions or advertisements.
In this example of multivariate analysis, the retail company is able
to distil the complexity of customer data into a smaller set of
meaningful components. This allows for a more focused and
effective approach to customer segmentation and marketing
strategies.
4. What are the ML techniques used for Multivariate analysis?
Several machine learning (ML) techniques can be employed for
multivariate analysis. Some common ones include:
A) Multiple Regression Analysis: Models the relationship
between a dependent variable and multiple independent
variables.
B) Principal Component Analysis (PCA): Reduces the
dimensionality of the data while preserving the most
important information, useful for identifying patterns in
multivariate datasets.
C)Factor Analysis: Examines the underlying structure among
observed variables, identifying common factors that explain
their correlations.
D)Canonical Correlation Analysis (CCA): Analyzes the
relationships between two sets of variables, finding linear
combinations that maximize the correlation between the
sets.
E) Cluster Analysis: Groups similar observations into clusters
based on the values of multiple variables, helping to identify
patterns in multivariate data.
F) Discriminant Analysis: Distinguishes between two or more
groups based on multiple continuous variables, useful for
classification problems.
G) Machine Learning Algorithms (e.g., Random Forest,
Support Vector Machines): Various ML algorithms can be
applied to multivariate datasets for tasks like classification,
regression, and clustering.
H)Neural Networks: Deep learning models, including neural
networks, can handle complex relationships among multiple
variables, making them suitable for intricate multivariate
analyses.
I) Manifold Learning Techniques (e.g., t-SNE, UMAP):
Visualizes high-dimensional data in lower dimensions,
aiding in the exploration of complex relationships within
multivariate datasets.
J) Ensemble Methods: Techniques like bagging and boosting
can enhance predictive performance in multivariate settings
by combining the outputs of multiple models.
These techniques offer diverse approaches to analyze and interpret
complex relationships in datasets with multiple variables. The
choice of method depends on the specific goals and
characteristics of the data.
