. What do you mean by Data Analysis?

Data analysis is a multidisciplinary field of data science, in which

data is analyzed using mathematical, statistical, and computer
science with domain expertise to discover useful information or
patterns from the data. It involves gathering, cleaning, transforming,
and organizing data to draw conclusions, forecast, and make
informed decisions. The purpose of data analysis is to turn raw data
into actionable knowledge that may be used to guide decisions,
solve issues, or reveal hidden trends.

2. How do data analysts differ from data scientists?

Data analysts and Data Scientists can be recognized by their

responsibilities, skill sets, and areas of expertise. Sometimes the
roles of data analysts and data scientists may conflict or not be
clear.
Data analysts are responsible for collecting, cleaning, and analyzing
data to help businesses make better decisions. They typically use
statistical analysis and visualization tools to identify trends and
patterns in data. Data analysts may also develop reports and
dashboards to communicate their findings to stakeholders.
Data scientists are responsible for creating and implementing
machine learning and statistical models on data. These models are
used to make predictions, automate jobs, and enhance business
processes. Data scientists are also well-versed in programming
languages and software engineering.

Feature Data analyst Data Scientist

Machine Learning,
Excel, SQL, Python, R, Statistical Modeling,
Tableau, PowerBI Docker, Software
Skills Engineering

Tasks Data Collection, Web Database

Scrapping, Data Management,
 Feature Data analyst Data Scientist

Predictive Analysis
and prescriptive
Cleaning, Data analysis, Machine
Visualization, Learning model
Explanatory Data building and
Analysis, Reports Deployment, Task
Development and automation, Work for
Presentations Business
Improvements
Process.

Positions Entry Label Seniors Label

3. How Data analysis is similar to Business

Intelligence?

Data analysis and Business intelligence are both closely related

fields, Both use data and make analysis to make better and more
effective decisions. However, there are some key differences
between the two.
 Data analysis involves data gathering, inspecting,
cleaning, transforming and finding relevant information, So,
that it can be used for the decision-making process.
 Business Intelligence(BI) also makes data analysis to find
insights as per the business requirements. It generally uses
statistical and Data visualization tools popularly known as BI
tools to present the data in user-friendly views like reports,
dashboards, charts and graphs.
The similarities and differences between the Data Analysis and
Business Intelligence are as follows:

Similarities Differences

Both use data to make better Data analysis is more technical,

decisions. while BI is more strategic.
 Similarities Differences

Data analysis focuses on finding

Both involve collecting, cleaning, patterns and insights in data,
and transforming data. while BI focuses on providing
relevant information

Data analysis is often used to

Both use visualization tools to provide specific answers, whereas
communicate findings. business intelligence (BI) is used
to help broader decision-making.

4. What are the different tools mainly used for data

analysis?

There are different tools used for data analysis. each has some
strengths and weaknesses. Some of the most commonly used tools
for data analysis are as follows:
 Spreadsheet Software: Spreadsheet Software is used for
a variety of data analysis tasks, such as sorting, filtering,
and summarizing data. It also has several built-in functions
for performing statistical analysis. The top 3 mostly used
Spreadsheet Software are as follows:
o Microsoft Excel
o Google Sheets
o LibreOffice Calc
 Database Management Systems (DBMS): DBMSs, or
database management systems, are crucial resources for
data analysis. It offers a secure and efficient way to manage,
store, and organize massive amounts of data.
o MySQL
o PostgreSQL
o Microsoft SQL Server
o Oracle Database
 Statistical Software: There are many statistical software
used for Data analysis, Each with its strengths and
weaknesses. Some of the most popular software used for
data analysis are as follows:
 o SAS: Widely used in various industries for statistical
analysis and data management.
o SPSS: A software suite used for statistical analysis
in social science research.
o Stata: A tool commonly used for managing,
analyzing, and graphing data in various [Link]:
 Programming Language: In data analysis, programming
languages are used for deep and customized analysis
according to mathematical and statistical concepts. For Data
analysis, two programming languages are highly popular:
o R: R is a free and open-source programming
language widely popular for data analysis. It has
good visualizations and environments mainly
designed for statistical analysis and data
visualization. It has a wide variety of packages for
performing different data analysis tasks.
o Python: Python is also a free and open-source
programming language used for Data analysis.
Nowadays, It is becoming widely popular among
researchers. Along with data analysis, It is used for
Machine Learning, Artificial Intelligence, and web
development.

5. What is Data Wrangling?

Data Wrangling is very much related concepts to Data

Preprocessing. It's also known as Data munging. It involves the
process of cleaning, transforming, and organizing the raw, messy or
unstructured data into a usable format. The main goal of data
wrangling is to improve the quality and structure of the dataset. So,
that it can be used for analysis, model building, and other data-
driven tasks.
Data wrangling can be a complicated and time-consuming process,
but it is critical for businesses that want to make data-driven
choices. Businesses can obtain significant insights about their
products, services, and bottom line by taking the effort to wrangle
their data.
Some of the most common tasks involved in data wrangling are as
follows:
  Data Cleaning: Identify and remove the errors,
inconsistencies, and missing values from the dataset.
 Data Transformation: Transformed the structure, format,
or values of data as per the requirements of the analysis.
that may include scaling & normalization, encoding
categorical values.
 Data Integration: Combined two or more datasets, if that
is scattered from multiple sources, and need of consolidated
analysis.
 Data Restructuring: Reorganize the data to make it more
suitable for analysis. In this case, data are reshaped to
different formats or new variables are created by
aggregating the features at different levels.
 Data Enrichment: Data are enriched by adding additional
relevant information, this may be external data or combined
aggregation of two or more features.
 Quality Assurance: In this case, we ensure that the data
meets certain quality standards and is fit for analysis.

6. What is the difference between descriptive and

predictive analysis?

Descriptive and predictive analysis are the two different ways to

analyze the data.
 Descriptive Analysis: Descriptive analysis is used to
describe questions like "What has happened in the past?"
and "What are the key characteristics of the data?". Its main
goal is to identify the patterns, trends, and relationships
within the data. It uses statistical measures, visualizations,
and exploratory data analysis techniques to gain insight into
the dataset.
The key characteristics of descriptive analysis are as follows:
o Historical Perspective: Descriptive analysis is
concerned with understanding past data and events.
o Summary Statistics: It often involves calculating
basic statistical measures like mean, median, mode,
standard deviation, and percentiles.
o Visualizations: Graphs, charts, histograms, and
other visual representations are used to illustrate
data patterns.
 o Patterns and Trends: Descriptive analysis helps
identify recurring patterns and trends within the
data.
o Exploration: It's used for initial data exploration and
hypothesis generation.
 Predictive Analysis: Predictive Analysis, on the other
hand, uses past data and applies statistical and machine
learning models to identify patterns and relationships and
make predictions about future events. Its primary goal is to
predict or forecast what is likely to happen in future.
The key characteristics of predictive analysis are as follows:
o Future Projection: Predictive analysis is used to
forecast and predict future events.
o Model Building: It involves developing and training
models using historical data to predict outcomes.
o Validation and Testing: Predictive models are
validated and tested using unseen data to assess
their accuracy.
o Feature Selection: Identifying relevant features
(variables) that influence the predicted outcome is
crucial.
o Decision Making: Predictive analysis supports
decision-making by providing insights into potential
outcomes.

7. What is univariate, bivariate, and multivariate

analysis?

Univariate, Bivariate and multivariate are the three different levels

of data analysis that are used to understand the data.
1. Univariate analysis: Univariate analysis analyzes one
variable at a time. Its main purpose is to understand the
distribution, measures of central tendency (mean, median,
and mode), measures of dispersion (range, variance, and
standard deviation), and graphical methods such as
histograms and box plots. It does not deal with the courses
or relationships from the other variables of the dataset.
Common techniques used in univariate analysis include
 histograms, bar charts, pie charts, box plots, and summary
statistics.
2. Bivariate analysis: Bivariate analysis involves the analysis
of the relationship between the two variables. Its primary
goal is to understand how one variable is related to the
other variables. It reveals, Are there any correlations
between the two variables, if yes then how strong the
correlations is? It can also be used to predict the value of
one variable from the value of another variable based on the
found relationship between the two.
Common techniques used in bivariate analysis include
scatter plots, correlation analysis, contingency tables, and
cross-tabulations.
3. Multivariate analysis: Multivariate analysis is used to
analyze the relationship between three or more variables
simultaneously. Its primary goal is to understand the
relationship among the multiple variables. It is used to
identify the patterns, clusters, and dependencies among the
several variables.
Common techniques used in multivariate analysis include
principal component analysis (PCA), factor analysis, cluster
analysis, and regression analysis involving multiple predictor
variables.

8. Name some of the most popular data analysis and

visualization tools used for data analysis.

Some of the most popular data analysis and visualization tools are
as follows:
 Tableau: Tableau is a powerful data visualization
application that enables users to generate interactive
dashboards and visualizations from a wide range of data
sources. It is a popular choice for businesses of all sizes
since it is simple to use and can be adjusted to match any
organization's demands.
 Power BI: Microsoft's Power BI is another well-known data
visualization tool. Power BI's versatility and connectivity with
other Microsoft products make it a popular data analysis and
visualization tool in both individual and enterprise contexts.
  Qlik Sense: Qlik Sense is a data visualization tool that is
well-known for its speed and performance. It enables users
to generate interactive dashboards and visualizations from
several data sources, and it can be used to examine
enormous datasets.
 SAS: A software suite used for advanced analytics,
multivariate analysis, and business intelligence.
 IBM SPSS: A statistical software for data analysis and
reporting.
 Google Data Studio: Google Data Studio is a free web-based
data visualization application that allows users to create
customized dashboards and simple reports. It aggregates
data from up to 12 different sources, including Google
Analytics, into an easy-to-modify, easy-to-share, and easy-
to-read report.

9. What are the steps you would take to analyze a

dataset?

Data analysis involves a series of steps that transform raw data into
relevant insights, conclusions, and actionable suggestions. While the
specific approach will vary based on the context and aims of the
study, here is an approximate outline of the processes commonly
followed in data analysis:
 Problem Definition or Objective: Make sure that the
problem or question you're attempting to answer is stated
clearly. Understand the analysis's aims and objectives to
direct your strategy.
 Data Collection: Collate relevant data from various
sources. This might include surveys, tests, databases, web
scraping, and other techniques. Make sure the data is
representative and [Link]
 Data Preprocessing or Data Cleaning: Raw data often
has errors, missing values, and inconsistencies. In Data
Preprocessing and Cleaning, we redefine the column's
names or values, standardize the formats, and deal with the
missing values.
 Exploratory Data Analysis (EDA): EDA is a crucial step in
Data analysis. In EDA, we apply various graphical and
 statistical approaches to systematically analyze and
summarize the main characteristics, patterns, and
relationships within a dataset. The primary objective behind
the EDA is to get a better knowledge of the data's structure,
identify probable abnormalities or outliers, and offer initial
insights that can guide further analysis.
 Data Visualizations: Data visualizations play a very
important role in data analysis. It provides visual
representation of complicated information and patterns in
the data which enhances the understanding of data and
helps in identifying the trends or patterns within a data. It
enables effective communication of insights to various
stakeholders.

10. What is data cleaning?

Data cleaning is the process of identifying the removing misleading

or inaccurate records from the datasets. The primary objective of
Data cleaning is to improve the quality of the data so that it can be
used for analysis and predictive model-building tasks. It is the next
process after the data collection and loading.
In Data cleaning, we fix a range of issues that are as follows:
1. Inconsistencies: Sometimes data stored are inconsistent
due to variations in formats, columns_name, data types, or
values naming conventions. Which creates difficulties while
aggregating and comparing. Before going for further
analysis, we correct all these inconsistencies and formatting
issues.
2. Duplicate entries: Duplicate records may biased analysis
results, resulting in exaggerated counts or incorrect
statistical summaries. So, we also remove it.
3. Missing Values: Some data points may be missing. Before
going further either we remove the entire rows or columns
or we fill the missing values with probable items.
4. Outlier: Outliers are data points that drastically differ from
the average which may result in machine error when
collecting the dataset. if it is not handled properly, it can
bias results even though it can offer useful insights. So, we
first detect the outlier and then remove it.
11. What is the importance of exploratory data
analysis (EDA) in data analysis?

Exploratory data analysis (EDA) is the process of investigating and

understanding the data through graphical and statistical techniques.
It is one of the crucial parts of data analysis that helps to identify
the patterns and trends in the data as well as help in understanding
the relationship between variables.
EDA is a non-parametric approach in data analysis, which means it
does take any assumptions about the dataset. EDA is important for
a number of reasons that are as follows:
1. With EDA we can get a deep understanding of patterns,
distributions, nature of data and relationship with another
variable in the dataset.
2. With EDA we can analyze the quality of the dataset by
making univariate analyses like the mean, median, mode,
quartile range, distribution plot etc and identify the patterns
and trends of single rows of the dataset.
3. With EDA we can also get the relationship between the two
or more variables by making bivariate or multivariate
analyses like regression, correlations, covariance, scatter
plot, line plot etc.
4. With EDA we can find out the most influential feature of the
dataset using correlations, covariance, and various bivariate
or multivariate plotting.
5. With EDA we can also identify the outliers using Box plots
and remove them further using a statistical approach.
EDA provides the groundwork for the entire data analysis process. It
enables analysts to make more informed judgments about data
processing, hypothesis testing, modelling, and interpretation,
resulting in more accurate and relevant insights.

12. What is Time Series analysis?

Time Series analysis is a statistical technique used to analyze and

interpret data points collected at specific time intervals. Time series
data is the data points recorded sequentially over time. The data
points can be numerical, categorical, or both. The objective of time
series analysis is to understand the underlying patterns, trends and
behaviours in the data as well as to make forecasts about future
values.
The key components of Time Series analysis are as follows:
 Trend: The data's long-term movement or direction over
time. Trends can be upward, downward, or flat.
 Seasonality: Patterns that repeat at regular intervals, such
as daily, monthly, or yearly cycles.
 Cyclical Patterns: Longer-term trends that are not as
regular as seasonality, and are frequently associated with
economic or business cycles.
 Irregular Fluctuations: Unpredictable and random data
fluctuations that cannot be explained by trends, seasonality,
or cycles.
 Auto-correlations: The link between a data point and its
prior values. It quantifies the degree of dependence
between observations at different time points.
Time series analysis approaches include a variety of techniques
including Descriptive analysis to identify trends, patterns, and
irregularities, smoothing techniques like moving averages or
exponential smoothing to reduce noise and highlight underlying
trends, Decompositions to separate the time series data into its
individual components and forecasting technique like ARIMA,
SARIMA, and Regression technique to predict the future values
based on the trends.

13. What is Feature Engineering?

Feature engineering is the process of selecting, transforming, and

creating features from raw data in order to build more effective and
accurate machine learning models. The primary goal of feature
engineering is to identify the most relevant features or create the
relevant features by combining two or more features using some
mathematical operations from the raw data so that it can be
effectively utilized for getting predictive analysis by machine
learning model.
The following are the key elements of feature engineering:
 Feature Selection: In this case we identify the most
relevant features from the dataset based on the correlation
with the target variables.
  Create new feature: In this case, we generate the new
features by aggregating or transforming the existing
features in such a way that it can be helpful to capture the
patterns or trends which is not revealed by the original
features.
 Transformation: In this case, we modify or scale the
features so, that it can helpful in building the machine
learning model. Some of the common transformations
method are Min-Max Scaling, Z-Score Normalization, and log
transformations etc.
 Feature encoding: Generally ML algorithms only process
the numerical data, so, that we need to encode categorical
features into the numerical vector. Some of the popular
encoding technique are One-Hot-Encoding, Ordinal label
encoding etc.

14. What is data normalization, and why is it

important?

Data normalization is the process of transforming numerical data

into standardised range. The objective of data normalization is scale
the different features (variables) of a dataset onto a common scale,
which make it easier to compare, analyze, and model the data. This
is particularly important when features have different units, scales,
or ranges because if we doesn't normalize then each feature has
different-different impact which can affect the performance of
various machine learning algorithms and statistical analyses.
Common normalization techniques are as follows:
 Min-Max Scaling: Scales the data to a range between 0
and 1 using the formula:
(x - min) / (max - min)
 Z-Score Normalization (Standardization): Scales data to
have a mean of 0 and a standard deviation of 1 using the
formula:
(x - mean) / standard_deviation
 Robust Scaling: Scales data by removing the median and
scaling to the interquartile range(IQR) to handle outliers
using the formula:
(X - Median) / IQR
  Unit Vector Scaling: Scales each data point to have a
Euclidean norm (length) (||X||) of 1 using the formula:
X / ||X||

15. What are the main libraries you would use for data
analysis in Python?

For data analysis in Python, many great libraries are used due to
their versatility, functionality, and ease of use. Some of the most
common libraries are as follows:
 NumPy: A core Python library for numerical computations.
It supports arrays, matrices, and a variety of mathematical
functions, making it a building block for many other data
analysis libraries.
 Pandas: A well-known data manipulation and analysis
library. It provides data structures (like as DataFrames) that
make to easily manipulate, filter, aggregate, and transform
data. Pandas is required when working with structured data.
 SciPy: SciPy is a scientific computing library. It offers a wide
range of statistical, mathematical, and scientific computing
functions.
 Matplotlib: Matplotlib is a library for plotting and
visualization. It provides a wide range of plotting functions,
making it easy to create beautiful and informative
visualizations.
 Seaborn: Seaborn is a library for statistical data
visualization. It builds on top of Matplotlib and provides a
more user-friendly interface for creating statistical plots.
 Scikit-learn: A powerful machine learning library. It
includes classification, regression, clustering, dimensionality
reduction, and model evaluation tools. Scikit-learn is well-
known for its consistent API and simplicity of use.
 Statsmodels: A statistical model estimation and
interpretation library. It covers a wide range of statistical
models, such as linear models and time series analysis.

16. What's the difference between structured and

unstructured data?
Structured and unstructured data depend on the format in which the
data is stored. Structured data is information that has been
structured in a certain format, such as a table or spreadsheet. This
facilitates searching, sorting, and analyzing. Unstructured data is
information that is not arranged in a certain format. This makes
searching, sorting, and analyzing more complex.
The differences between the structured and unstructured data are
as follows:

Feature Structured Data Unstructured Data

Schema (structure of
No predefined
data) is often rigid
relationships between
and organized into
data elements.
Structure of data rows and columns

Excellent for
searching, reporting, Difficult to search
Searchability and querying

Simple to quantify No fixed format,

and process using making it more
standard database challenging to
Analysis functions. organize and analyze.

Storage Relational databases Data lakes

Customer records,
Text documents,
product inventories,
images, audio, video
Examples financial data

17. How can pandas be used for data analysis?

Pandas is one of the most widely used Python libraries for data
analysis. It has powerful tools and data structure which is very
helpful in analyzing and processing data. Some of the most useful
functions of pandas which are used for various tasks involved in
data analysis are as follows:
 1. Data loading functions: Pandas provides different
functions to read the dataset from the different-different
formats like read_csv, read_excel, and read_sql functions
are used to read the dataset from CSV, Excel, and SQL
datasets respectively in a pandas DataFrame.
2. Data Exploration: Pandas provides functions
like head, tail, and sample to rapidly inspect the data after it
has been imported. In order to learn more about the
different data types, missing values, and summary statistics,
use pandas .info and .describe functions.
3. Data Cleaning: Pandas offers functions for dealing with
missing values (fillna), duplicate rows (drop_duplicates), and
incorrect data types (astype) before analysis.
4. Data Transformation: Pandas may be used to modify and
transform data. It is simple to do actions like selecting
columns, filtering rows (loc, iloc), and adding new ones.
Custom transformations are feasible using
the apply and map functions.
5. Data Aggregation: With the help of pandas, we can group
the data using groupby function, and also apply aggregation
tasks like sum, mean, count, etc., on specify columns.
6. Time Series Analysis: Pandas offers robust support for
time series data. We can easily conduct date-based
computations using functions like resample, shift etc.
7. Merging and Joining: Data from different sources can be
combined using Pandas merge and join functions.

18. What is the difference between pandas Series and

pandas DataFrames?

In pandas, Both Series and Dataframes are the fundamental data

structures for handling and analyzing tabular data. However, they
have distinct characteristics and use cases.
A series in pandas is a one-dimensional labelled array that can hold
data of various types like integer, float, string etc. It is similar to a
NumPy array, except it has an index that may be used to access the
data. The index can be any type of object, such as a string, a
number, or a datetime.
A pandas DataFrame is a two-dimensional labelled data structure
resembling a table or a spreadsheet. It consists of rows and
columns, where each column can have a different data type. A
DataFrame may be thought of as a collection of Series, where each
column is a Series with the same index.
The key differences between the pandas Series and Dataframes are
as follows:

pandas Series pandas DataFrames

A one-dimensional labelled array

A two-dimensional labelled data
that can hold data of various
structure that resembles a table or
types like (integer, float, string,
a spreadsheet.
etc.)

Similar to a spreadsheet, which

Similar to the single vector or
can have multiple vectors or
column in a spreadsheet
columns as well as.

The versatility and handling of the

Best suited for working with
multiple features make it suitable
single-feature data
for tasks like data analysis.

DataFrames can be assumed as a

Each element of the Series is
collection of multiple Series,
associated with its label known as
where each column shares the
the index
same index.

19. What is One-Hot-Encoding?

One-hot encoding is a technique used for converting categorical

data into a format that machine learning algorithms can understand.
Categorical data is data that is categorized into different groups,
such as colors, nations, or zip codes. Because machine learning
algorithms often require numerical input, categorical data is
represented as a sequence of binary values using one-hot encoding.
To one-hot encode a categorical variable, we generate a new binary
variable for each potential value of the category variable. For
example, if the category variable is "color" and the potential values
are "red," "green," and "blue," then three additional binary variables
are created: "color_red," "color_green," and "color_blue." Each of
these binary variables would have a value of 1 if the matching
category value was present and 0 if it was not.

20. What is a boxplot and how it's useful in data

science?

A boxplot is a graphic representation of data that shows the

distribution of the data. It is a standardized method of the
distribution of a data set based on its five-number summary of data
points: the minimum, first quartile [Q1], median, third quartile [Q3],
and maximum.

Boxplot
Boxplot is used for detection the outliers in the dataset by
visualizing the distribution of data.

Statistics Interview Questions and Answers

for Data Analyst
21. What is the difference between descriptive and
inferential statistics?

Descriptive statistics and inferential statistics are the two main

branches of statistics
 Descriptive Statistics: Descriptive statistics is the branch
of statistics, which is used to summarize and describe the
main characteristics of a dataset. It provides a clear and
concise summary of the data's central tendency, variability,
and distribution. Descriptive statistics help to understand
the basic properties of data, identifying patterns and
structure of the dataset without making any generalizations
beyond the observed data. Descriptive statistics compute
measures of central tendency and dispersion and also
create graphical representations of data, such as
histograms, bar charts, and pie charts to gain insight into a
dataset.
Descriptive statistics is used to answer the following
questions:
o What is the mean salary of a data analyst?
o What is the range of income of data analysts?
o What is the distribution of monthly incomes of data
analysts?
 Inferential Statistics: Inferential statistics is the branch of
statistics, that is used to conclude, make predictions, and
 generalize findings from a sample to a larger population. It
makes inferences and hypotheses about the entire
population based on the information gained from a
representative sample. Inferential statistics use hypothesis
testing, confidence intervals, and regression analysis to
make inferences about a population.
Inferential statistics is used to answer the following
questions:
o Is there any difference in the monthly income of the
Data analyst and the Data Scientist?
o Is there any relationship between income and
education level?
o Can we predict someone's salary based on their
experience?

22. What are measures of central tendency?

Measures of central tendency are the statistical measures that

represent the centre of the data set. It reveals where the majority of
the data points generally cluster. The three most common measures
of central tendency are:
 Mean: The mean, also known as the average, is calculated
by adding up all the values in a dataset and then dividing by
the total number of values. It is sensitive to outliers since a
single extreme number can have a large impact on the
mean.
Mean = (Sum of all values) / (Total number of values)
 Median: The median is the middle value in a data set when it
is arranged in ascending or descending order. If there is an
even number of values, the median is the average of the
two middle values.
 Mode: The mode is the value that appears most frequently
in a dataset. A dataset can have no mode (if all values are
unique) or multiple modes (if multiple values have the same
highest frequency). The mode is useful for categorical data
and discrete distributions.

23. What are the Measures of dispersion?

Measures of dispersion, also known as measures of variability or
spread, indicate how much the values in a dataset deviate from the
central tendency. They help in quantifying how far the data points
vary from the average value.
Some of the common Measures of dispersion are as follows:
 Range: The range is the difference between the highest and
lowest values in a data set. It gives an idea of how much the
data spreads from the minimum to the maximum.
 Variance: The variance is the average of the squared
deviations of each data point from the mean. It is a measure
of how spread out the data is around the mean.
Variance(σ2)=∑(X−μ)2NVariance(σ2)=N∑(X−μ)2
 Standard Deviation: The standard deviation is the square
root of the variance. It is a measure of how spread out the
data is around the mean, but it is expressed in the same
units as the data itself.
 Mean Absolute Deviation (MAD): MAD is the average of
the absolute differences between each data point and the
mean. Unlike variance, it doesn't involve squaring the
differences, making it less sensitive to extreme values. it is
less sensitive to outliers than the variance or standard
deviation.
 Percentiles: Percentiles are statistical values that measure
the relative positions of values within a dataset. Which is
computed by arranging the dataset in descending order
from least to the largest and then dividing it into 100 equal
parts. In other words, a percentile tells you what percentage
of data points are below or equal to a specific value.
Percentiles are often used to understand the distribution of
data and to identify values that are above or below a certain
threshold within a dataset.
 Interquartile Range (IQR): The interquartile range (IQR) is
the range of values ranging from the 25th percentile (first
quartile) to the 75th percentile (third quartile). It measures
the spread of the middle 50% of the data and is less
affected by outliers.
 Coefficient of Variation (CV): The coefficient of variation
(CV) is a measure of relative variability, It is the ratio of the
standard deviation to the mean, expressed as a percentage.
 It's used to compare the relative variability between
datasets with different units or scales.

24. What is a probability distribution?

A probability distribution is a mathematical function that estimates

the probability of different possible outcomes or events occurring in
a random experiment or process. It is a mathematical
representation of random phenomena in terms of sample
space and event probability, which helps us understand the relative
possibility of each outcome occurring.
There are two main types of probability distributions:
1. Discrete Probability Distribution: In a discrete
probability distribution, the random variable can only take
on distinct, separate values. Each value is associated with a
probability. Examples of discrete probability distributions
include the binomial distribution, the Poisson distribution,
and the hypergeometric distribution.
2. Continuous Probability Distribution: In a continuous
probability distribution, the random variable can take any
value within a certain range. These distributions are
described by probability density functions (PDFs). Examples
of continuous probability distributions include the normal
distribution, the exponential distribution, and the uniform
distribution.

25. What are normal distributions?

A normal distribution, also known as a Gaussian distribution, is a

specific type of probability distribution with a symmetric, bell-
shaped curve. The data in a normal distribution clustered around a
central value i.e mean, and the majority of the data falls within one
standard deviation of the mean. The curve gradually tapers off
towards both tails, showing that extreme values are becoming
distribution having a mean equal to 0 and standard deviation equal
to 1 is known as standard normal distribution and Z-scores are used
to measure how many standard deviations a particular data point is
from the mean in standard normal distribution.
Normal distributions ar