Computational Statistics

Computational Statistics - Insem Exam Questions (2022, 2023, 2024)

Q.No Question Marks

Unit 1 - Introduction to Statistics:- 7 hours:

What is statistics, Statistical Data- Categorical, Numerical (Continuous), Uni-variate and Bi-variate
Analysis, Mean, Median, Mode, Standard Deviation, Harmonic Mean, Data Visualization-Line,
Scatter, Box plots, Histogram, Statistical Thinking.

1 What is statistics and their types? Explain importance of statistics. 7

https://www.youtube.com/watch?v=PB0qtNNeNIM

https://www.youtube.com/watch?v=7V5jtI-ihm0

Statistics is a field of study that focuses on collecting, organizing, analyzing,

interpreting, and presenting data in a clear and meaningful way. It allows us to make
sense of raw information by identifying patterns, trends, and insights.
In both everyday life and professional fields like science, business, and healthcare,
statistics helps us solve problems and make informed decisions based on data. It plays a
role from the planning phase—such as designing surveys or experiments—to interpreting
results.
Types of Statistics
Statistics is generally divided into two main branches:
1. Descriptive Statistics
Descriptive statistics help summarize and describe the main features of a dataset. Rather
than making predictions, it focuses on what the data tells us directly.
Common tools include averages, percentages, medians, ranges, and graphs.
Example: Reporting the average age of students in a classroom.
2. Inferential Statistics
Inferential statistics go a step further by using sample data to draw conclusions or make
predictions about a larger population.
This process involves hypothesis testing, estimating parameters, and using
probability theory.

1
 It allows us to test theories and make assumptions about a broader group without
studying every single member.
Example: Studying a sample of 100 voters to predict the behavior of an entire
city’s electorate.
Key Statistical Terminologies
Population: The entire group you want to study (e.g., all students in a country).
Sample: A small subset taken from the population for analysis (e.g., 200 students
from different schools).
Variable: A measurable trait or characteristic (e.g., age, income, score).
Distribution: Describes how data is spread across different values or ranges.
Parameter: A numerical value that describes a feature of the population (e.g.,
population mean).
Quantitative Analysis: Focuses on numerical data (e.g., mean, variance).
Qualitative Analysis: Focuses on non-numerical aspects (e.g., data type,
organization, or quality).
Importance of Statistics
Statistics is vital because it:
Supports decision-making in fields like healthcare, business, education, and
government.
Simplifies complex data, turning raw numbers into clear summaries.
Enables prediction and forecasting, such as estimating future sales or disease
outbreaks.
Guides scientific research by testing hypotheses and drawing valid conclusions.
Helps detect patterns, trends, and relationships among variables.
Inferential Statistics Procedure (Step-by-Step)
Start with a theory: Begin with a general concept or idea.
Form a research hypothesis: Make a specific, testable prediction (e.g., “More
sleep improves grades”).
Define variables: Decide what you’ll measure (e.g., sleep hours and exam
scores).
Identify the population: Clarify who or what you’re studying.

2
 Set the null hypothesis: Assume there’s no effect (e.g., sleep has no impact on
grades).
Collect a sample: Choose a representative group from the population.
Conduct statistical tests: Use methods (like t-tests or chi-square tests) to check if
the results are statistically significant.

2 Explain Mean, Medium and Mode with examples. 8

https://www.youtube.com/watch?v=keLk7odKCsE

https://www.youtube.com/watch?v=sc0OCzvKCwA

https://www.youtube.com/watch?v=X48cZ6DGaSw

Mean, Medium and Mode are main measures of central tendency, to describe the center
or typical value of a dataset.

1. Mean (Average)

The mean represents the sum of all values divided by the total count of values.

Formula: Mean = Sum of all values / Number of values

Example: Heights of 5 students: 150, 160, 155, 165, 170

Mean = (150 + 160 + 155 + 165 + 170) ÷ 5 = 800 ÷ 5 = 160

Therefore, the mean height is 160 cm.

2. Median (Middle Value)

3
 The median represents the central value when data is arranged in order. For datasets
with an even number of values, the median equals the average of the two central
numbers.

Example: Heights: 150, 160, 155, 165, 170

→ Arranged: 150, 155, 160, 165, 170 The central value is 160, making the median 160
cm.

For even counts: 150, 155, 160, 165 → Arranged = 150, 155, 160, 165

Median = (155 + 160) ÷ 2 = 157.5 cm

3. Mode (Most Frequent Value)

The mode identifies the value that occurs most frequently in a dataset. Datasets may
contain one mode, multiple modes, or no mode.

Example: Heights: 150, 160, 160, 165, 170 Since 160 appears twice while other values
appear once, the mode is 160 cm.

When all values occur equally: No mode exists.

3 Consider the following data points: 17, 16, 21, 18, 15, 17, 21, 19, 11, 23. 7
Calculate the mean, median and mode of the above data points.
1. Mean
The mean is the average of all the data points.
 Sum of data points: 17+16+21+18+15+17+21+19+11+23=178
 Number of data points: 10
 Mean: 10178=17.8
2. Median
The median is the middle value when the data points are arranged in ascending order.
 Arrange data in ascending order: 11,15,16,17,17,18,19,21,21,23
 Since there are an even number of data points (10), the median is the average of

4
 the two middle values (the 5th and 6th values).
 Middle values: 17 and 18
 Median: 217+18=235=17.5
3. Mode
The mode is the value(s) that appear most frequently in the data set.
 Count occurrences of each value:
o 11: 1
o 15: 1
o 16: 1
o 17: 2
o 18: 1
o 19: 1
o 21: 2
o 23: 1
 Mode: Both 17 and 21 appear twice, which is more than any other value.
Therefore, this data set has two modes (it is bimodal).
4 What is Dispersion? Explain any two types of Dispersion with example. 6

https://www.youtube.com/watch?v=64ELhoTvzk0

https://www.youtube.com/watch?v=pRZdoMeB9Xo

https://www.youtube.com/watch?v=Ug2s0Qr3ijU

5
In statistics, dispersion (also known as variability, scatter, or spread) refers to the extent
to which a distribution of data is stretched or squeezed. In simpler terms, it quantifies
how much individual data points differ from the central tendency (like the mean, median,
or mode) of a dataset.
Common measures of dispersion include:

 Range: The difference between the highest and lowest values.

 Interquartile Range (IQR): The range of the middle 50% of the data.
 Variance: The average of the squared differences from the mean.
 Standard Deviation: The square root of the variance, giving a measure of spread
in the original units of the data.
 Mean Absolute Deviation (MAD): The average of the absolute differences from
the mean.

Explain any two:

1. Range
The Range is the simplest measure of dispersion. It represents the difference between the
highest (maximum) and lowest (minimum) values in a dataset.
Formula: Range = Maximum Value - Minimum Value
Explanation: The range provides a quick and easy way to understand the total spread of
the data. It gives you an idea of the "span" of your dataset from one extreme to the other.
However, its simplicity is also its limitation: it only uses two data points and is highly
sensitive to outliers. A single extreme value can significantly inflate the range, making it
not always representative of the typical spread of the data.
Example: Imagine you are a teacher recording the scores of 10 students on a recent
quiz: Student scores: 12,18,15,20,11,16,19,13,17,25
To calculate the range:
1. Identify the highest score (Maximum Value): The highest score is 25.
2. Identify the lowest score (Minimum Value): The lowest score is 11.
3. Calculate the Range: Range = Maximum Value - Minimum Value Range =
25−11=14

6
 Interpretation: The range of 14 indicates that the scores on the quiz span 14 points, from
the lowest score of 11 to the highest score of 25. This gives us a quick idea of the
variability in student performance.
2. Variance
The Variance is a more sophisticated measure of dispersion that quantifies how much
individual data points deviate from the mean (average) of the dataset. It considers all data
points and provides a more comprehensive understanding of the data's spread.
Formula (for a population):

 σ2 (sigma squared) is the population variance

 Xi is each individual data point
 μ (mu) is the population mean
 N is the total number of data points in the population

5 Differentiate between Univariate, Bivariate and Multivariate Data Analysis. 9

https://www.youtube.com/watch?v=_beKM1XkYTk

https://www.youtube.com/watch?v=JG8GRlMjp3c

https://www.youtube.com/watch?v=HNYSiOlsQQY

https://www.youtube.com/watch?v=Ep0HuvMxnAA

Parameter Univariate Analysis Bivariate Analysis Multivariate Analysis

Analysis of one Analysis of two Analysis of three or

Definition
variable variables more variables

To understand the To study the

To explore the
distribution or relationship
Objective relationships among
pattern of one between two
multiple variables
variable variables

7
 Three or more columns
Data Involved Single column of data Two columns of data
of data

Multiple regression,
Techniques Mean, median, mode, Correlation, scatter
PCA, clustering,
Used histograms, box plots plots, regression line
heatmaps

Studying the relation Predicting house price

Example Finding the average
between height and using area, number of
Scenario age of students
weight rooms, and location

Graph 3D plots, pair plots,

Bar chart, histogram Scatter plot
Example heatmaps

Sample Area, Bedrooms,

Age: [15, 16, 15, 17, Height vs Weight of
Dataset Location, Price for each
18] students
Example house

Univariate Analysis
 One variable
 Focus: Distribution, central tendency
 Tools: Mean, Histogram, Boxplot
 Example: Analyzing students' heights
Univariate – Just one list of values.
Bivariate Analysis
 Two variables
 Focus: Relationship between variables
 Tools: Correlation, Scatter Plot, Linear Regression
 Example: Study hours vs. exam scores
Bivariate – A table with 2 columns like:
Height (cm) | Weight (kg)

------------|-------------

160 | 55

165 | 60

8
Multivariate Analysis
 Three or more variables
 Focus: Complex relationships and patterns
 Tools: Multiple Regression, PCA, Clustering
 Example: Predicting house price using size, location, etc.
Multivariate – A table with 3+ columns:

Area (sqft) | Bedrooms | Location | Price ($)

------------|----------|----------|----------

1200 | 3 | Urban | 200,000

Mathematical Equations
Univariate Analysis
You're analyzing just one column of data. Most common goal: central tendency or
spread.
Mean (Average):
x̄ = (1/n) ∑ xᵢ

Variance:
σ² = (1/n) ∑ (xᵢ - x̄)²
Standard Deviation:
σ = √σ²

Example: For ages [15, 16, 15, 17, 18],

Mean = (15 + 16 + 15 + 17 + 18) / 5 = 16.2

Bivariate Analysis
Analyzing how two variables relate to each other.
Covariance:
Cov(X, Y) = (1/n) ∑ (xᵢ - x̄)(yᵢ - ȳ)
Correlation (Pearson’s r):
r = Cov(X, Y) / (σₓ * σᵧ)
Simple Linear Regression:
Y = a + bX
where b = Cov(X, Y) / Var(X), a = ȳ - b * x̄
Multivariate Analysis
Analyzing multiple variables to predict or understand patterns.
Multiple Linear Regression:
Y = a + b₁X₁ + b₂X₂ + ... + bₖXₖ

9
 Covariance Matrix:
Σᵢⱼ = Cov(Xᵢ, Xⱼ)
Example:
Price = 50,000 + 100 × Area + 10,000 × Rooms + 5,000 × Location Score
6 Explain Importance of Data Visualization and explain scatter plot. 8

Data Visualization

https://www.youtube.com/watch?v=cbonyxuX0_U&t=56s

https://www.youtube.com/watch?v=leXt9btuyEY

https://www.youtube.com/watch?v=tkui0ohdZaM

https://www.youtube.com/watch?v=hLbVXF70BCE

Scatter plot

https://www.youtube.com/watch?v=aWmHYYQJukk

https://www.youtube.com/watch?v=5GihhMCPucg

https://www.youtube.com/watch?v=AxY7nTkgR6M

Data Visualization is the graphical presentation of data using elements such as charts,
graphs, maps, and infographics to make information easier to understand, analyze and
communicate. It transforms raw data into visual formats that make it easier to:
 Understand complex data quickly
 Reveal trends and patterns
 Detect anomalies and outliers
 Make informed decisions faster

Importance of Data
Explanation Example
Visualization
Converts large, complex
1. Simplifies Complex Using a bar chart to show
datasets into easily
Data monthly sales trends
understandable visual formats
Helps identify trends, patterns,
2. Reveals Patterns and Line graphs showing stock
and outliers that may not be
Trends price changes over time
obvious in raw data
3. Improves Decision Enables faster and better- Dashboard showing key KPIs
Making informed decisions by for business decisions

10
 presenting insights visually
Makes it easier to share insights
4. Enhances Pie charts for market share
and findings with others clearly
Communication presentations
and quickly
Visuals are more attractive and
Infographics summarizing
5. Engages Audience engaging than plain text or
survey results
tables
Interactive visuals allow users to
6. Supports Data Interactive maps showing
explore data from different
Exploration sales by region
angles
7. Identifies Scatter plots illustrating
Visual tools help detect
Relationships and correlation between
relationships between variables
Correlations advertising spend and revenue
8. Facilitates Real-time dashboards help track
Real-time sales dashboard in
Monitoring and performance and spot issues
retail management
Tracking quickly
Scatter Plot

A scatter plot is a basic yet powerful graphical tool used in statistics and data science to
display individual data points on a two-dimensional plane. Each point corresponds to
paired values from two different variables. To represent more variables, variations in
color, size, or shape of the markers can be used.
When to Use a Scatter Plot
Scatter plots are especially helpful for:
 Investigating the relationship or correlation between two continuous variables
 Detecting patterns, clusters, or groups within the data
 Identifying outliers or unusual data points that deviate from the overall trend

11
 Importance of Scatter Plot
Scatter plots are important because they provide a clear visual representation of how two
variables interact, allowing analysts to quickly understand correlations, spot natural
groupings, and identify exceptions. This insight aids in making informed decisions and
guides further statistical analysis.
7 What is Histogram? How do you construct Histogram? 7
Differentiate between Bar Graph and Histogram.

12
A histogram is a graphical representation that displays the frequency distribution of
continuous numerical data. It divides the data into continuous intervals called bins, and
each bar’s height represents the number of data points that fall within that specific
interval.
Example:

Height Range (cm) Number of Students

140–150 3
150–160 7
160–170 12
170–180 6

What is a Bar Graph?

A bar graph (or bar chart) is a visual representation used to compare different categories
of data using rectangular bars. Each bar’s length or height corresponds to the value or
frequency of that category. Bars are separated by spaces to emphasize that the categories
are distinct.

Example
Fruit Number of People
Apple 10
Banana 15
Orange 5

13
 Difference between Bar Graph and Histogram

Feature Bar Graph Histogram

A graph that represents categorical A graph that represents continuous
Definition
data using bars. numerical data using bars.
Type of Data Categorical (qualitative) data Quantitative (continuous) data
Bar Spacing Bars are separated by gaps Bars are adjacent (no gaps)
Width of
Can vary or remain the same Usually equal width
Bars
X-axis Represents categories Represents class intervals (bins)
Y-axis Represents frequency or count Represents frequency or density
Bars are in increasing order of class
Arrangement Bars can be in any order
intervals
Graph showing favourite fruit Distribution of students’ heights in
Example
choices: Apples, Bananas, etc. cm (140–150, 150–160, etc.)
- Comparing different products - Analysing distributions
Applications - Survey results - Statistical analysis
- Business sales by category - Quality control
Used to compare different Used to understand distribution of
Interpretation
categories numerical data
8 What is box plot and its uses? 7
How will you differentiate between histogram and box plot.
A box plot is a graphical summary of a set of numerical data, highlighting its central
value, spread, and potential outliers using five-number summary statistics:

14
 Element Description
Minimum The smallest value (excluding outliers)
Q1 (Lower Quartile) The 25th percentile — 25% of the data falls below this value
Median (Q2) The 50th percentile — the middle value of the dataset
Q3 (Upper Quartile) The 75th percentile — 75% of the data falls below this value
Maximum The largest value (excluding outliers)
IQR (Interquartile
Q3 − Q1: Measures the spread of the middle 50% of data
Range)
Lines that extend from the box to the minimum and maximum
Whiskers
(non-outlier) values
Points that fall outside 1.5×IQR from Q1 or Q3, shown as
Outliers
individual dots

Example: # Numerical example data: test scores of 15 students

scores = [55, 60, 65, 67, 68, 70, 72, 74, 76, 78, 80, 85, 88, 90, 100]

Sample

15
Uses of Box Plot:

Purpose Explanation
Clearly shows how spread out the data is, especially the
Visualizing Data Spread
central 50%
Identifying Outliers Highlights extreme values that may skew the analysis
Multiple box plots can be used side-by-side to compare
Comparing Distributions
datasets
Detecting Skewness Uneven whiskers or off-centre median indicate data skew
Understanding Central Median line shows central location without being affected
Tendency by outliers
Offers a compact overview of distribution without needing
Quick Summary of Data
to look at raw data

Differentiate between histogram and box plot

Feature Histogram Box Plot (Box-and-Whisker Plot)

Definition A graphical representation of A graphical summary of data

16
 the distribution of numerical showing the median, quartiles, and
data using bars to show outliers using a five-number
frequency within intervals summary.
(bins).
Shows the shape of the data
Summarizes data spread, central
Purpose distribution (e.g., normal,
tendency, and detects outliers.
skewed).
Bars touching each other; A box with whiskers, a line for the
Visual Structure
height = frequency. median, and dots for outliers.
Key Minimum, Q1, Median (Q2), Q3,
Bins (intervals), frequencies.
Components Maximum, Outliers, IQR.
Data Type Continuous numerical data. Continuous numerical data.
Shows
No (only summarizes spread and
Distribution Yes (e.g., bell-shaped, skewed).
central tendency).
Shape
Yes (displayed as dots beyond
Outliers Visible? No (not clearly).
whiskers).
Understanding the frequency Comparing multiple datasets quickly
Ideal For
and pattern of data values. or checking for variability & outliers.
Comparison of Easier to compare multiple datasets
Harder to compare side by side.
Multiple Sets side by side.

Example Use Case Histogram Example Box Plot Example

Shows how many students scored Summarizes min, Q1, median,
Analysing exam
in each range (50–60, 60–70, Q3, max, and outliers in one
score distribution
etc.) view
Monitoring Reveals peak time ranges or long Highlights central tendency and
customer wait times queues any unusually long waits
Measuring heights Visualizes the shape of height Detects outliers (very short or
of students distribution very tall students)
9 Describe relation between statistics and machine learning. 8

https://www.youtube.com/watch?v=TW8GDs6oFzk

https://www.youtube.com/watch?v=ld60jtk-TI0

https://www.youtube.com/watch?v=QOGhtUGqP94&t=1s

Statistics:
The science of collecting, analyzing, interpreting, and presenting data to support

17
decision-making and uncover patterns.
Machine Learning:
A subset of artificial intelligence that enables systems to learn from data and improve
their performance on tasks without being explicitly programmed.
[Clarity for “Without being explicitly programmed”: the system is not given specific
step-by-step instructions for every possible scenario. Instead, it learns patterns and rules
from data to make decisions or predictions on its own.
Example:
In traditional programming, we write rules to detect spam emails.
In machine learning, you train a model with examples of spam and non-spam emails and
it learns to classify new emails without manually coding the rules]
Relation between Statistics and Machine Learning
Statistics forms the foundation of many processes in machine learning. It helps us
understand data, prepare it for modeling, and evaluate model performance.
Key statistical tasks in machine learning include:
 Data preprocessing: Detecting outliers, handling missing values, scaling
features, encoding categorical data, and sampling subsets of data.
 Descriptive statistics: Summarizing data using measures like mean, median,
mode, variance, and standard deviation. These help identify patterns and
relationships in the data (univariate, bivariate, or multivariate analysis).
 Model evaluation: Statistics provides metrics such as accuracy, precision,
recall, F1-score, and confidence intervals to assess how well a machine learning
model performs, especially on unseen data.
Statistical Data Types Used in ML
 Quantitative Data (Numerical): Includes measurable variables like price, speed,
and ratings. Used for algorithms like regression and clustering.
 Qualitative Data (Categorical): Includes variables grouped into categories like
gender or brand. Encoded numerically for classification algorithms.
o Nominal: No inherent order (e.g., colors, brands).
o Ordinal: Ordered categories (e.g., grades, size).
Why Statistics is Essential in ML

18
  Helps in choosing the right model for a given problem.
 Ensures data quality and reliability through cleaning and exploration.
 Provides tools to interpret results and explain model decisions.
 Aids in avoiding misleading conclusions by understanding variance, bias, and
overfitting.
Example
Suppose you're analyzing leisure activities using survey data. Statistics helps you:
 Understand how often each activity is done (distribution),
 Find average participation (central tendency),
 Observe variation in responses (variability).
Machine learning can then use this structured data to predict future behavior or segment
users using clustering or classification.
Summary
Statistics is not just a support tool but a core element of machine learning. It enables
effective data handling, informed decision-making, and model validation—making
statistical knowledge essential for every ML practitioner.
Comparison:

Points Statistics Machine Learning

Data collection, analysis, and Learning patterns and making
Primary Focus
inference using formal models predictions from data
Modelling Assumes specific distributions; Often assumption-free; models
Approach emphasizes mathematical rigor discovered from data
Hypothesis testing, confidence Algorithms (e.g., decision trees,
Key Techniques
intervals, p-values neural networks, SVMs)
Terminology Parameters, covariates Weights, features
Data Excels with large, high-
Effective for small datasets
Requirements dimensional data
Requires explicit assumptions for Minimal assumptions; more
Assumptions
validity flexible
Explanation, inference,
Goal Orientation Prediction accuracy, automation
interpretability
Learning Iterative training and
Static model estimation
Process optimization
High—models are often Varies—some models (e.g., deep
Interpretability
explainable and statistically learning) are less interpretable

19
 validated
Performance P-values, confidence intervals, R², Accuracy, precision, recall, F1-
Metrics etc. score, AUC
Rooted in mathematics and Rooted in computer science and
Origin
probability (centuries old) AI (modern evolution)
Increasing use of ML techniques to Adoption of statistical principles
Integration
augment traditional statistical for robustness, especially on
Trend
methods small data
Converging fields—mutual Toward a unified approach to
Future Outlook borrowing of methods and data analysis and problem-
practices solving

20