Foundations of Data Science

By Dr. Shagun Sharma

 Table of Contents
What is Data Science? Big Data and Data Science
Datafication – Current landscape of perspectives – Skill sets needed;
Matrices – Matrices to represent relations between data, and necessary linear
algebraic operations on matrices
Approximately representing matrices by decompositions (SVD and PCA);
Statistics: Descriptive Statistics: distributions
and probability – Statistical Inference: Populations and samples – Statistical
modeling – probability distributions – fitting a model – Hypothesis Testing –
Intro to R/ Python.
 What is Data Science?
● Data Science is a multidisciplinary field that uses scientific methods, algorithms, and
tools to extract knowledge and insights from large, complex datasets (structured &
unstructured) to solve problems, find patterns, and make predictions, blending
statistics, AI, and computer science for data-driven decisions in areas like business,
healthcare, and tech.
● It involves collecting, cleaning, analyzing, visualizing, and interpreting data to
understand "what happened," "why," and "what might happen next," ultimately guiding
better actions.
What is Data Science?
Find the Structured and Unstructured Data
Find the Structured and Unstructured Data
 What does data science find?

● What do customers want?

● How can we improve our services?
● What will the upcoming trend in sales?
● How much stock they need for upcoming festival.
Data science empowers the industries to make smarter,
faster, and more informed decisions. In order to find
patterns and achieve such insights, expertise in relevant
domain is required. With expertise in Healthcare, a data
scientists can predict patient risks and suggest
personalized treatments.
 Real-world examples of data science

● Netflix/Spotify: Suggesting what you might like next based on your viewing/listening
history.
● E-commerce: Showing you relevant products and personalizing offers.
● Healthcare: Predicting disease outbreaks or understanding treatment effectiveness.
● Finance: Detecting fraud or forecasting stock market trends.
 Where data science is being used?

Data Science is being used in almost all major industry. Here are some examples:

● Predicting customer preferences for personalized recommendations.

● Detecting fraud in financial transactions.
● Forecasting sales and market trends.
● Enhancing healthcare with predictive diagnostics and personalized treatments.
● Identifying risks and opportunities in investments.
● Optimizing supply chains and inventory management.
 Data Science Life Cycle

Data science is not a one-step process such that you will get to learn it in a short time
and call ourselves a Data Scientist. It's passes from many stages and every element is
important. One should always follow the proper steps to reach the ladder. Every step
has its value and it counts in your model.
 Data Science Life Cycle

1. Problem Statement: Identifying and defining the problem.

2. Data Collection: After defining the problem statement, the next step is to go in
search of data that you might require for your model. You must do good research, find
all that you need. Data can be in any form i.e unstructured or structured. It might be in
various forms like videos, spreadsheets, coded forms, etc. You must collect all these
kinds of sources.
3. Data Cleaning: Data cleaning is all about the removal of missing, redundant,
unnecessary and duplicate data from your collection.
4. Data Analysis and Exploration: It analyzes the structure of data, finding hidden
patterns in them, studying behaviors, visualizing the effects of one variable over others
and then concluding.
 Data Science Life Cycle

5. Data Modelling: In this step, we choose an algorithm that best fits to the data. There
different kinds of algorithms from regression, classification, and Clustering, etc.
6. Optimization and Deployment: In this step, you test your data and find how well it is
performing by checking its accuracy. The efficiency of the model is checked and thus try
to optimize it for better accurate prediction. Deployment deals with the launch of the
model and let the people outside there to benefit from that. You can also obtain
feedback from organizations and people to know their need and then to work more on
your model.
 Data Science Tools and Libraries

● Jupyter Notebook: Interactive environment for coding and documentation.

● Google Colab: Cloud-based Jupyter Notebook for collaborative coding.
● TensorFlow: Deep learning framework for building neural networks.
● PyTorch: Popular library for machine learning and deep learning.
● Scikit-learn: Tools for predictive data analysis and machine learning.
● Docker: Containerization for reproducible environments.
● Kubernetes: Managing and scaling containerized applications.
● Apache Kafka: Real-time data streaming and processing.
● Tableau: A powerful tool for creating interactive and shareable data visualizations.
● Power BI: A business intelligence tool for visualizing data and generating insights.
● Keras: A user-friendly library for designing and training deep learning models.
 What is Big Data?
● Big Data is a massive collection of complex information (text, numbers, images,
videos) that's too large and fast for normal tools like spreadsheets to handle, requiring
special systems to find valuable patterns for better decisions, like predicting customer
needs or improving services.
● It's defined by the 5 Vs that are:
1. Volume
2. Velocity
3. Variety
4. Veracity
5. Value
What is Big Data?
 5 V’s in Big Data

1. Volume: The sheer massive scale of data being generated from sources like social
media, sensors, and transactions, often too large for traditional systems.
2. Velocity: The speed at which data is created, collected, and needs to be processed,
such as real-time stock trades or streaming data.
3. Variety: The diversity of data formats, including structured (databases), unstructured
(text, images, video), and semi-structured data.
 5 V’s in Big Data

1. Volume: The sheer massive scale of data being generated from sources like social
media, sensors, and transactions, often too large for traditional systems.
2. Velocity: The speed at which data is created, collected, and needs to be processed,
such as real-time stock trades or streaming data.
3. Variety: The diversity of data formats, including structured (databases), unstructured
(text, images, video), and semi-structured data.
4. Veracity: The accuracy, reliability, and trustworthiness of the data, dealing with
inconsistencies and biases to ensure data quality.
5. Value: The ultimate goal which is extracting meaningful, actionable insights and
economic benefit from the data to drive decisions and innovation.
Datafication
 What is Datafication?

● Datafication is the process of converting aspects of life, business, and society—like

behaviors, interactions, and phenomena—into quantifiable data that can be collected,
stored, and analyzed, providing raw material for data science, machine learning, and
AI to reveal patterns, make predictions, and generate new value, moving beyond
simple digitization to capture the "what" and "how" of activities.
● It turns previously unmeasurable things (like customer sentiment or traffic flow) into
digital records, enabling deeper insights and improved decision-making.
How Datafication is Performed?
 How Datafication is Performed?

1. Data Collection: It all starts with data collection and retrieval. This could be from
anything you do—clicking on a website, using an app, or even just walking around with
your smartphone. Devices and sensors collect this data, often without you even noticing.
2. Data Storage: Once collected, this data needs a place to go. Think of it like storing
your favorite movies or music. The data is saved in databases or cloud storage, where it
can be accessed and used later.
3. Data Processing: Here’s where things get interesting. The raw data collected isn’t
very useful on its own. It’s like having all the ingredients for a cake but not baking it yet.
Data processing involves data cleaning, organizing, and transforming this data into a
more usable format. For example, if you’ve ever used a spreadsheet to track your
expenses, you’ve engaged in a basic form of data processing.
 How Datafication is Performed?

4. Data Analysis: This is the magic moment when data becomes valuable information.
Using various tools and techniques, analysts can look at patterns and trends in the data.
For example, they might discover that people tend to buy more ice cream on hot
days—a useful insight for a business.
5. Data Visualization: To make the data easy to understand, it’s often presented
visually, like in charts or graphs. If you’ve ever seen a bar chart showing monthly sales
or a line graph tracking your steps over time, you’ve encountered data visualization.
This step helps people quickly grasp the insights hidden in the data.
6. Data Application: Finally, the insights gained from data analysis are put to use. This
could mean anything from tweaking a marketing strategy to designing a new product.
For example, if data shows that customers prefer shopping online at certain times of the
day, a business might run targeted ads during those hours.
Why Datafication is important?
 How Datafication is Performed?

1. Informed Decision-Making
Imagine you’re planning a vacation. You want to find the best hotel at the best price,
right? Datafication plays a huge role here. By analyzing data from various travel
websites, you can compare prices, read reviews, and even see trends in booking rates.
2. Personalization
Have you ever noticed how streaming or retail services like Netflix or Amazon seem to
know exactly what you want to watch or order next? That’s datafication at work.
3. Efficiency and Innovation
Datafication isn’t just about making things convenient; it’s also about driving efficiency
and innovation. Think about the last time you used a navigation app like Google Maps.
The app collects real-time traffic data from millions of users to provide the quickest route
to your destination.
 How Datafication is Performed?

4. Enhanced User Experience

Companies use datafication to improve your overall experience with their products or
services. For instance, if you’ve ever received an email from a company asking for
feedback on a recent purchase, that’s datafication at work.
5. Predictive Capabilities
Another exciting aspect of datafication is its predictive capabilities. By analyzing past
behaviors and trends, businesses can predict future outcomes.
For example, a retailer might use data to forecast which products will be popular during
the holiday season, allowing them to stock up accordingly.
Real World Examples of Datafication
 Real World Examples of Datafication

1. Social Media
These platforms collect data on your interactions—likes, shares, comments, and even
the time you spend looking at certain posts. By analyzing this data, social media
companies can tailor your feed to show more of what you like and less of what you
don’t.

This isn’t just about keeping you engaged, it’s also about delivering targeted ads that are
relevant to you. If you’ve ever seen an ad pop up for something you were just thinking
about, it’s because datafication has been at work, analyzing your behavior and
preferences.
 Real World Examples of Datafication

2. Smart Homes
Imagine coming home after a long day, and your house adjusts the lighting,
temperature, and even plays your favorite music as you walk in.

Devices like smart thermostats, lights, and security systems collect data on your daily
routines and preferences. They learn when you typically get home, your preferred
temperature settings, and even the times when you’re usually away.

This data helps automate tasks, making your life more convenient and energy-efficient.
It’s like having a personal butler who knows your preferences and schedules.
 Real World Examples of Datafication

3. Fitness Trackers
If you’ve ever used a fitness tracker like a Fitbit or an Apple Watch, you’re already
familiar with datafication in action. These devices collect data on your steps, heart rate,
sleep patterns, and more.
This data isn’t just for show, it helps you understand your health and fitness levels. For
example, by tracking your steps and calories burned, you can set and achieve fitness
goals.
If your heart rate spikes unexpectedly, your device can alert you to potential health
issues. Moreover, many fitness apps allow you to share your data with healthcare
providers, giving them valuable insights into your health that can lead to better, more
personalized care.
 Real World Examples of Datafication

4. Retail and Online Shopping

Have you ever noticed how online stores like Amazon seem to know exactly what you
want to buy? This isn’t just clever marketing, it’s datafication at work.

Retailers track your browsing history, past purchases, and even the items you’ve looked
at but didn’t buy. By analyzing this data, they can recommend products that are tailored
to your tastes and needs.

This personalized shopping experience not only makes it easier for you to find what
you’re looking for but also introduces you to new products you might not have
considered otherwise.
 Real World Examples of Datafication

5. Navigation and Ride-Sharing Apps

Ever used Google Maps or a ride-sharing app like Uber or Ola? These services are
excellent examples of datafication in action. They collect data from millions of users to
provide real-time traffic updates, optimal routes, and estimated arrival times.
For ride-sharing apps, this data helps match you with drivers and calculate fare
estimates based on distance, traffic, and time of day. This not only makes your commute
more efficient but also enhances safety by providing you with accurate and up-to-date
information.
These examples show how datafication is seamlessly integrated into our daily lives,
often in ways we don’t even notice. As you continue to interact with these technologies,
being aware of how your data is used can help you make more informed decisions and
fully enjoy the benefits of a data-driven world.
 Current landscape of
perspectives – Skill sets
needed
 Current landscape of perspectives

Business/Economic
Focuses on driving efficiency, customer understanding, and new revenue streams
through data-driven decisions.
Technical/Analytical
Emphasizes using statistical methods, machine learning (ML), and AI to find patterns
and predict outcomes.
Ethical/Societal
Addresses privacy, bias, and the responsible use of data, ensuring fairness and
minimizing negative impacts.
Interdisciplinary
Recognizes the need for collaboration between data scientists, domain experts
(healthcare, finance), and humanities to tackle complex problems.
 Current landscape of perspectives

AI and Machine Learning

Data science is increasingly intertwined with artificial intelligence (AI) and machine
learning, enabling more advanced predictive analytics and automation of data-driven
processes.
Big Data
The proliferation of big data continues to shape the data science landscape, with
organizations leveraging large volumes of data from diverse sources to gain insights
and make informed decisions.
Ethics and Privacy
There is a growing emphasis on ethical considerations and privacy concerns in data
science, with a focus on responsible data usage, transparency, and compliance with
regulations such as GDPR.
 Current landscape of perspectives

Interdisciplinary Collaboration
Data science is increasingly viewed as a multidisciplinary field, requiring collaboration
between data scientists, domain experts, and stakeholders to ensure the meaningful
interpretation and application of data-driven insights.
Data Visualization and Interpretability
The importance of effective data visualization and interpretability techniques is on the
rise, as stakeholders seek to comprehend and communicate complex data findings in a
more accessible manner.
 Skill sets needed

Programming
Python (for data manipulation/ML) & SQL (for database queries).
Data Engineering
ETL (Extract, Transform, Load) tools (Apache Airflow, Talend), cloud platforms (AWS
Glue), building data pipelines.
Data Analysis & ML
Statistics, probability, regression models, machine learning algorithms (PCA, SVD), and
AI.
Data Visualization
Tools to present insights clearly (dashboards, graphs).
 Skill sets needed

Domain Knowledge
Understanding the specific industry (finance, healthcare) to ask the right questions and
interpret results.
Soft Skills
Critical thinking, problem-solving, and ethical awareness to handle data responsibly.
 Matrices
– Matrices to represent
relations between data
 What is Matrices in Data Science?

● Matrices are a foundational concept in data science that underpins a wide range of
mathematical and computational operations used for analyzing and manipulating
data.
● It provides a structured and organized way to represent information, making it easier
to process and extract meaningful insights. In this comprehensive explanation, we’ll
delve deeper into matrices in the context of data science, exploring their properties,
operations, and applications.
 What is Matrices in Data Science?

Matrix Basics
● A matrix is a two-dimensional array of numbers arranged in rows and columns. Each
element in a matrix is identified by its row and column index.
● A matrix with “m” rows and “n” columns is often referred to as an “m x n” matrix.
Matrices are used to represent datasets, where each row corresponds to an
observation or sample, and each column represents a feature or attribute of that
sample.
● This structured representation makes it convenient to apply mathematical operations
and transformations to the data.
 What is Matrices in Data Science?

Data Representation
● In data science, matrices serve as a powerful tool for representing datasets. Consider
a dataset containing information about various individuals, such as age, income, and
education level.
● By organizing this data into a matrix, where each row corresponds to an individual
and each column represents a different attribute, we create a structured
representation that facilitates analysis.
● This tabular arrangement simplifies operations like finding averages, and correlations,
and performing statistical analyses.
 What is Matrices in Data Science?

Linear Transformation
● Matrices are key players in the realm of linear transformations, which are fundamental
to data manipulation and feature engineering.
● These transformations involve scaling, rotating, reflecting, and translating data points.
In data science, linear transformations are utilized for data preprocessing and
dimensionality reduction.
● For example, Principal Component Analysis (PCA) leverages matrices to identify
orthogonal axes that maximize the variance in data, leading to effective dimensionality
reduction.
 What is Matrices in Data Science?

Matrix Operation
Matrices support a multitude of operations that are essential in data science:
● Addition and subtraction: Matrices with the same dimensions can be added or
subtracted element-wise, facilitating tasks such as aggregating data from multiple
sources.
● Scalar Multiplication: Each element of a matrix can be multiplied by a scalar value,
which can be useful for scaling data.
● Matrix Multiplication: Matrix multiplication is a central operation that combines the
rows and columns of matrices to produce a new matrix. The element at position (i, j)
in the resulting matrix is the dot product of the “i”-th row of the first matrix and the “j”-th
column of the second matrix. Matrix multiplication is crucial for composing linear
transformations and forms the foundation of various machine learning algorithms.
 What is Matrices in Data Science?

Matrix Operation
Matrices support a multitude of operations that are essential in data science:
● Transpose: The transpose of a matrix is obtained by interchanging its rows and
columns. This operation is valuable for solving systems of linear equations and for
extracting features in certain algorithms.
 What is Matrices in Data Science?

Image and Signal Processing

In image and signal processing, matrices are used to represent images and signals as
pixel values in a grid. Operations like convolution are applied to matrices to perform
tasks such as edge detection and feature extraction in images. Convolutional Neural
Networks (CNNs) use matrix convolutions to learn and recognize patterns in images.

Graphs and Networks

Matrices are used to represent relationships in graphs and networks. The adjacency
matrix, for instance, represents connections between nodes in a graph. Matrices like the
Laplacian matrix help analyze graph properties and identify clusters or communities
within networks.
 What is Matrices in Data Science?

● Matrices are a cornerstone of data science, facilitating the representation,

transformation, and analysis of data.
● Understanding matrices and their operations is vital for proficiently applying machine
learning algorithms, conducting statistical analyses, and extracting insights from
complex datasets.
● By harnessing the power of matrices, data scientists can unlock the potential hidden
within data and drive informed decision-making across various domains.
Necessary linear algebraic
operations on matrices
 What is Matrix?

● A Matrix is a rectangular arrangement of numbers in rows and columns.

● In a matrix, as we know rows are the ones that run horizontally and columns are the
ones that run vertically. The following image shows the examples of matrices.
 What are the Algebraic Operations?

● Basic algebraic operations are any one of the traditional operations of arithmetic,
which are addition, subtraction, multiplication, division, raising to an integer power,
and taking roots. These operations may be performed on numbers, in which case they
are often called arithmetic operations.
 Singular Value
Decomposition (SVD)
 SVD

● Singular Value Decomposition (SVD) is a powerful mathematical technique in data

science used to break down a dataset (represented as a matrix or a table of numbers)
into three simpler, meaningful components. This process helps uncover hidden
patterns, simplify data, and reduce noise.
● SVD (Singular Value Decomposition) in data science is a powerful matrix factorization
T
technique that breaks a data matrix (A) into three simpler matrices (U), (∑), (V ),
revealing underlying patterns for tasks like dimensionality reduction, noise reduction,
data compression, and recommendation systems. It identifies key features (singular
vectors in (U) and (V) and their importance (singular values in (∑) by extracting
dominant correlations, allowing for simplification and better interpretation of
high-dimensional data.
 SVD

● Singular Value Decomposition (SVD) is a factorization method in linear algebra that

decomposes a matrix into three other matrices, providing a way to represent data in
terms of its singular values.

SVD helps you split that table into three parts:

● U: This part tells you about the people (like their general preferences).
● Σ: This part shows how important each factor is (how much each rating matters).
● Vᵀ: This part tells you about the products (how similar they are to each other)
 Example of SVD

● SVD breaks this table into three smaller parts: one that shows people’s preferences,
one that shows the importance of each movie, and one that shows how similar the
movies are to each other.
● Mathematically, the SVD of a matrix A (size m × n) is represented as: A = U Σ VT
 Example of SVD

● SVD breaks this table into three smaller parts: one that shows people’s preferences,
one that shows the importance of each movie, and one that shows how similar the
movies are to each other.
● Mathematically, the SVD of a matrix A (size m × n) is represented as: A = U Σ VT
 Why SVD?

● Dimensionality Reduction: SVD can reduce the number of features (columns) in a

large dataset while retaining the most important information. This makes data easier
to work with and speeds up machine learning algorithms (similar to how Principal
Component Analysis, or PCA, works).

● Image Compression: SVD is used to compress large images efficiently. By keeping

only a few of the largest singular values and their corresponding vectors, you can
reconstruct an image that looks almost identical to the human eye but uses
significantly less storage space.
 Why SVD?

● Recommendation Systems: Platforms like Netflix and Amazon use SVD to analyze
user-item interaction data. It helps uncover hidden preferences (latent factors) and
predict what a user might like to watch or buy next, leading to personalized
recommendations.

● Natural Language Processing (NLP): SVD is used in a technique called Latent

Semantic Analysis (LSA) to understand the relationships between words and
documents, improving search engine relevance and topic modeling.
 Example of SVD?

● Take a Matrix

● Goal of SVD:
A=UΣVT
Where
U: left singular vectors (orthogonal)
Σ: singular values (diagonal, ≥ 0)
V: right singular vectors (orthogonal)
 Example of SVD?

● STEP 1: Compute ATA

● STEP 2: Find Eigenvalues of ATA

 Example of SVD?

● STEP 3: Compute Singular Values

● STEP 4: Form Σ
Example of SVD?
Example of SVD?
Example of SVD?
Step by step Breakdown
Step by step Breakdown
Step by step Breakdown
Step by step Breakdown
Step by step Breakdown
 PCA (Principle Component Analysis)

● In simple terms, Principal Component Analysis (PCA) is a smart way to simplify

complex data by reducing its many features (dimensions) into a smaller, more
manageable set of "principal components" that still hold almost all the important
information, like summarizing a detailed book into a few key chapters.
● It finds the most important directions (components) in the data where the spread
(variance) is greatest, making data faster for models to learn from, easier to visualize,
and helping to remove noise or redundancy.
● PCA (Principal Component Analysis) is a dimensionality reduction technique and
helps us to reduce the number of features in a dataset while keeping the most
important information. It changes complex datasets by transforming correlated
features into a smaller set of uncorrelated components.
 How PCA Works?

● Imagine you have a detailed photo of a cat with tons of pixels (features).
● PCA finds the best angles to look at the cat so you can describe it with fewer words
(components) while still capturing its essential shape, eyes, and ears, discarding
irrelevant background details.
 Key goals of PCA

● Dimensionality Reduction: Takes many features (like height, weight, age, income) and
combines them into fewer, new features (PCs).
● Information Preservation: Captures the most significant patterns and variance in the
data, ensuring important info isn't lost.
● Noise Reduction: Filters out random noise, leading to cleaner data.
● Faster Models: Simpler data means machine learning algorithms train quicker and
perform better
 When it's used

● Data Visualization: Makes high-dimensional data viewable on a 2D or 3D graph.

● Preprocessing: A common first step before applying other ML algorithms to speed
them up.
● Feature Extraction: Creating new, uncorrelated features from existing ones.
 SVD Vs PCA

Singular Value Decomposition (SVD)

● What it does: A matrix factorization technique that breaks down any matrix (not just square
or symmetric ones) into three simpler matrices (U, Σ, Vᵀ).
● Goal: Decompose data into fundamental components (rotations, scaling, another rotation).
● How: Gives you singular vectors (directions) and singular values (importance/magnitude).
Principal Component Analysis (PCA)
● What it does: Finds new, uncorrelated variables (Principal Components) that capture the
maximum variance in the data, helping to reduce dimensions and visualize complex
datasets.
● Goal: Feature extraction and dimensionality reduction by identifying the most significant
patterns (directions of highest variance).
● How: Finds eigenvectors and eigenvalues of the covariance matrix.
PCA Solved Example
PCA Solved Example
PCA Solved Example
PCA Solved Example
 SVD Vs PCA

Singular Value Decomposition (SVD)

● What it does: A matrix factorization technique that breaks down any matrix (not just square
or symmetric ones) into three simpler matrices (U, Σ, Vᵀ).
● Goal: Decompose data into fundamental components (rotations, scaling, another rotation).
● How: Gives you singular vectors (directions) and singular values (importance/magnitude).
Principal Component Analysis (PCA)
● What it does: Finds new, uncorrelated variables (Principal Components) that capture the
maximum variance in the data, helping to reduce dimensions and visualize complex
datasets.
● Goal: Feature extraction and dimensionality reduction by identifying the most significant
patterns (directions of highest variance).
● How: Finds eigenvectors and eigenvalues of the covariance matrix.
Statistics: Descriptive Statistics &
Statistical Inference
Detailed Explanation with Diagrams
 Descriptive Statistics
• Descriptive statistics summarizes data.
– Measures of Central Tendency: Mean, Median, Mode.
– Measures of Dispersion: Variance, Standard Deviation.
– Used for understanding dataset characteristics.
 Data Distributions
• A distribution shows how data values are spread.
– Symmetric vs Skewed distributions.
– Discrete vs Continuous distributions.
 Normal Distribution

Bell-shaped symmetric distribution. Mean = Median = Mode. 68-95-99.7 Rule

applies.
 Binomial Distribution

Discrete probability distribution. Fixed number of trials. Two outcomes per trial.
 Statistical Inference
• Used to draw conclusions about population using
sample data.
– Includes estimation and hypothesis testing.
– Based on probability theory.
 Population vs Sample
• Population: Entire group of interest.
– Sample: Subset selected from population.
– Sample statistics estimate population parameters.
 Sampling Distribution (Central Limit
Theorem)

Even if population is uniform, sampling distribution of mean approaches normal

distribution.
 Statistical Modeling
• Model represents real-world data mathematically.
– Example: Linear Regression, Logistic Regression.
– Helps in prediction and inference.
 Fitting a Model
• Choose appropriate probability distribution.
– Estimate parameters (Mean, Variance).
– Evaluate goodness of fit (Chi-square, KS test).
 Hypothesis Testing
• Null Hypothesis (H0) vs Alternative Hypothesis
(H1).
– Choose significance level α.
– Compute test statistic and p-value.
– Decision: Reject or Fail to Reject H0.
 Hypothesis Testing Curve

Critical regions at ±1.96 for α = 0.05 (two-tailed Z-test).

THANK YOU