MID II ASSIGNMENT

BUSINESS STATISTICS (BUM-DSM-123)

TOPIC: BUSINESS FORECASTING
MARKETING ANALYTICS

by

Name: Deepti Dangi

Roll no.: Y24282006

Under the guidance of

Professor Lokesh Uke Sir

DEPARTMENT OF MASTERS OF BUSINESS ADMINISTRATION

Dr. Hari Singh Gour Central University of Sagar

 BUSINESS FORECASTING

Business Forecasting is a broad term that refers to business forecasting

techniques through the development of sophisticated models. These forecasting
models help predict the numerous business developments that can happen in the
near future which helps the business leaders make better decisions and avoid
potential pitfalls. (Bansal, 2020)

Forecasting is a whole different phenomenon that refers to foreseeing how

some value or event would turn out to be over a period of time. Here, the past
data is analysed at the backdrop of time as the role of time is also taken into
account to come up with the forecasted value.

While all businesses do some form of forecasting or other, the level of their
business forecasting techniques can differ substantially. While a simple analysis
can be done of the past data to understand the pattern and forecast the future of
the business or economic events, other sophisticated statistical models can be
put to use to know of the future. With the advancement of Machine Learning
and Deep Learning, algorithms working under such an advanced setup can also
be used.
 There are mainly two types of models- Qualitative Models and
Quantitative Models.

Qualitative Model
This forecasting model is used when trying to predict business important
prediction over a short period of time. These business forecasting methods also
are commonly used when there is a lack of historical data or the future seems
too unstable to forecast using strictly a mathematical or statistical tool. While
there are various methods that come under this model, the two most common
business forecasting methods include Market Research and Delphi Model.

Quantitative Model
The quantitative method solely relies on the historical data and understands the
underlying pattern to know the future. This model is used when there is no
dearth of historical data and there is a need to forecasting not only for the short
term but medium or even long term. Among the Quantitative Model, there are a
number of commonly used business forecasting techniques such as- Time Series
Analysis, Econometric Method, Indicator Approach.
 • Business Development
Once the basics of forecasting are developed, the next important element is to
estimate the development of the business. This includes knowing the steps that
the business is taking in the upcoming time to achieve some goals that the
business leaders have in mind and the course of action the business is going to
take. In addition to this, the upcoming conditions and plausible events that can
happen in the concerned industry are estimated. This estimation is done based
on experienced experts’ insight (qualitative model) or the analysis of the
historical data (quantitative model) or both.

• Tuning of Model
This element focuses on the evaluation of the forecasting model by comparing
the forecast done for a historical timeframe and comparing it with the actual
values. Here the error needs to be quantified and then minimized by
understanding the reasons for the error. These errors or deviations form the basis
to fine-tune the model, take anomalies/outliers into account, and use different
approaches until that method is figured out, which gives the best result.

• Implementing of Model
Once the model is finalized, the model is to be implemented to perform
forecasting. On the basis of the forecasted values, business strategies are to be
formulated and policies are to be driven. All of this must be done to keep in
mind the error that can be there between the forecasted and actual values. The
crucial element here is the perpetual improvement that is to be made in the data
collection, model implementation, and other processes to maximize the
accuracy of the forecasted values.
 • Short Term Forecasting

This forecasting level is generally done for up to 1 year (and is often less than 3
months). A business that faces challenges such as inventory management and
needs to know the number of products to purchase this forecasting type. This
idea can be extended to forecasting workforce requirements, job assignments,
etc.

• Medium-Term Forecasting

This range is generally considered between 1 year (or 3 months) to 3 years.

Business leaders often need to forecast sales, demand, profit, or loss to form
strategies, budgets, and other things, which is where forecasting techniques
come in handy.

• Long Term Forecasting

Forecasting is done over 3 year’s time span is considered as Long Term

Forecasting. Here economic forecasting is often done such as understanding
Macro Economic, which includes knowing of inflation, job requirements,
money supply, etc. Knowing all this can help governments and other big entities
 TIME SERIES FORECASTING
Time series forecasting is a statistical technique used to analyze data points
gathered at consistent intervals over a time span in order to detect patterns and
trends. Understanding the fundamental framework of the data can assist in
predicting future data points and making knowledgeable choices. (Wainaina,
2023)

• To understand how time series works and what factors affect a certain
variable(s) at different points in time.

• Time series analysis will provide the consequences and insights of the
given dataset’s features that change over time.

• Supporting to derive the predicting the future values of the time series
variable.

With the help of “Time Series,” we can prepare numerous time-based analyses
and results.

• Forecasting: Predicting any value for the future.

• Segmentation: Grouping similar items together.

• Classification: Classifying a set of items into given classes.

• Descriptive analysis: Analysis of a given dataset to find out what is there

in it.
• Intervention analysis: Effect of changing a given variable on the
outcome.

Components of Time Series

• Trend: In which there is no fixed interval and any divergence within the
given dataset is a continuous timeline. The trend would be Negative or
Positive or Null Trend

• Seasonality: In which regular or fixed interval shifts within the dataset in

a continuous timeline. Would be bell curve or saw tooth

• Cyclical: In which there is no fixed interval, uncertainty in movement

and its pattern

• Irregularity: Unexpected situations/events/scenarios and spikes in a

short time span.
Time series has the below-mentioned limitations; we must take care of those
during our data analysis.

• Similar to other models, the missing values are not supported by TSA

• The data points must be linear in their relationship.

• Data transformations are mandatory, so they are a little expensive.

• Models mostly work on Uni-variate data.

Time series models are statistical tools that experts use to study and predict data
that changes over time. These models help us uncover patterns, trends, and
relationships in the data, which in turn allows us to make informed predictions
about what might happen in the future.

Below is a brief overview of some commonly used time series models:

Moving Average (MA) Model: This model calculates the average of past
observations with the aim of predicting future values. It is useful for capturing
short-term fluctuations and random variations in the data.

• Assumptions: The observations are a linear combination of past error

terms, and there is no autocorrelation between the error terms.

• Parameters: The order of the model (q) determines the number of lagged
error terms to include.

• Strengths: MA models are effective in capturing short-term

dependencies and smoothing out random fluctuations in the data.
Autoregressive Moving Average (ARMA) Model: The ARMA model
combines the AR and MA models to capture both short-term and long-term
patterns in the data. It is effective for analyzing stationary time series data.

• Assumptions: The observations are a linear combination of past

observations and past error terms, and there is no autocorrelation between
the error terms.

• Parameters: The orders of the AR and MA components (p and q)

determine the number of lagged observations and error terms to include.

• Strengths: ARMA models combine the strengths of AR and MA models,

capturing both short-term and long-term dependencies in the data.

Autoregressive Integrated Moving Average (ARIMA) Model: This model

extends the ARMA model by incorporating differencing to handle non-
stationary data. It is suitable for data with trends or seasonality.

• Assumptions: The data is stationary after differencing, meaning the

differences between consecutive observations are stationary.

• Parameters: The orders of the AR, I, and MA components (p, d, and q)

determine the number of lagged observations, differencing, and lagged
error terms to include.

• Strengths: ARIMA models can handle non-stationary data by

incorporating differencing, making them suitable for time series with
trends or seasonality.
 LIFETIME VALUE CALCULATION
The process by which a business measures the value of a customer to the
business through the customer’s full lifespan
Lifetime Value Calculation is the process by which a business measures the
value of a customer to the business through the customer’s full lifespan.
Customer Lifetime Value or LTV is one of the metrics used to measure the
growth of a company. (Ali, 2024)

By comparing the LTV of a company to the cost of customer acquisition, it can

calculate the value of a customer to the business over the period of time that
they were associated with them. The LTV helps a company gain and retain
highly valuable customers.
 What is Customer Lifetime Value

The customer lifetime value (LTV), also known as lifetime value, is the
total revenue a company expects to earn over the lifetime of their relationship
with a single customer. The customer lifetime value calculation accounts for the
customer acquisition costs, operating expenses, and costs to produce the goods
or services that the company is manufacturing. Many companies tend to
overlook the LTV metric but the lifetime value of customers is essential to the
growth of a company.

The lifetime value calculation is given as:

How to Calculate the LTV of a

• Average purchase value – It is calculated by dividing the company’s

total revenue over a period of time by the total purchases made by its
customers during that same timeframe.

• Average purchase frequency rate – It is calculated by the total

purchases made over a period of time by the individual customers that
made those purchases during that time.
 • Customer value – It is calculated by multiplying the average value of the
purchase by the number of times the purchase is made.

• Average customer lifespan – It is the average number of years that a

customer continues to buy the company’s goods and services.

• Lifetime value calculation – The LTV is calculated by multiplying the

value of the customer to the business by their average lifespan. It helps a
company identify how much revenue they can expect to earn from a
customer over the life of their relationship with the company.

What Factors Contribute to Lifetime

The lifetime value of a business depends on how popular the brand is among
customers. For example, if a customer lacks any loyalty to the brand and does
not face any switching costs when buying a rival company’s product, it can
result in a negative impact on the lifetime value of the company. Following are
the factors that affect the LTV of a company:

1. Churn rate

The churn rate describes how often customers stop shopping at a business that
they were once loyal customers of. The rate can differ from business to business
and depends on the competitive advantage of the company and their ability to
keep customers interested in their products. Usually, small businesses and
startups tend to face a high churn rate.

2. Brand loyalty
It measures how loyal the customers are to the brand and who keep buying their
goods and services. Building brand loyalty can help retain customers and
decrease the churn rate. A company with a lot of loyal customers will generate a
high lifetime value.
There are many tactics that businesses can implement to boost efficiency and
increase customer retention rates, thereby increasing their LTV:

1. Good communication

Open communication between the business and the customer can help a
customer relate to the brand better. It is important for companies to listen to
feedback from their customers as it can help them improve and grow. Effective
communication reduces the churn rate as well.

2. Re-engage customers

An effective way to increase LTV is to engage with customers who previously

purchased goods and services from the company. It is particularly useful for
companies with a long shelf life, and it can help improve brand recognition.

3. Increase brand loyalty

The lifetime value of a company can help with future growth projections and
increase profitability. LTV can be increased by implementing strategies to
increase brand loyalty.
 CREDIT SCORING MODEL
Credit scoring models are statistical tools that evaluate creditworthiness and
determine the likelihood of default on credit obligations. These models are used
by credit bureaus and lenders to assess the risk of lending money or extending
credit to individuals or businesses.
A credit score typically ranges from 300 to 850, with a higher score indicating a
lower risk of default. Lenders use credit scores to make decisions about loan
terms, including interest rates, repayment periods, and loan amounts. A good
credit score can result in favorable loan terms, while a poor score can lead to
higher interest rates and less favorable terms. (jhonson, 2023)

The credit risk scoring model provides a standardized and objective way for
lenders to assess the creditworthiness of individuals and businesses. By using a
credit scoring model, lenders can evaluate the risk of lending money or
extending credit to a borrower, allowing them to make informed decisions about
loan terms and interest rates.
Without a credit risk scoring model, lenders would have to rely on subjective
judgments and personal opinions when evaluating a borrower’s
creditworthiness. This could result in inconsistencies and potentially
discriminatory lending practices. A standardized credit scoring model ensures
that all borrowers are evaluated based on the same criteria, creating a fair and
transparent lending process.

There are various types of credit scoring models used in finance, each with its
own unique methodology and criteria. Understanding the different types of
credit scoring models can help individuals and businesses make informed
decisions about credit and loans.
 1. FICO Score:
The FICO score is the most commonly used credit scoring model in the United
States. It uses a range of factors to calculate a credit score, including payment
history, credit utilization, length of credit history, types of credit accounts, and
recent credit inquiries. FICO scores range from 300 to 850, with higher scores
indicating a lower risk of default.
Here is a look at each category and the weight it carries in determining the
credit score:
• Payment history (35%): This factor evaluates how consistently a
borrower has made payments on their debts. A borrower who has always
made on-time payments will receive a higher score than one who has
missed payments.
• Credit utilization (30%): This factor evaluates the percentage of
available credit that’s being used. A borrower who uses less than 30% of
their available credit will receive a higher score than one who uses more.
• Length of credit history (15%): This factor evaluates how long a
borrower has had credit accounts open. A borrower who has a long
history of credit accounts in good standing will receive a higher score
than one who is new to credit.

2. VantageScore:
The VantageScore is a newer credit scoring model that was developed jointly by
the three major credit bureaus. It also uses a range of factors to calculate a credit
score, but weighs them differently than the FICO score. VantageScores range
from 300 to 850, with higher scores indicating a lower risk of default.
VantageScore 4.0, the latest version of the model, uses six factors to calculate a
credit score: payment history, age and type of credit, percentage of credit limit
used, total balances and debt, recent credit behavior and inquiries, and available
credit. The VantageScore model puts less emphasis on payment history and
more emphasis on credit utilization than the FICO model.Here is a look at each
category and the weight it carries in determining the credit score:
• Payment history (40%): This factor evaluates how consistently a
borrower has made payments on their debts, similar to the FICO score.
 • Age and type of credit (21%): This factor evaluates the borrower’s
credit history, including the age of their oldest and newest credit accounts
and the mix of credit types.
• Percentage of credit limit used (20%): This factor evaluates the
borrower’s credit utilization, similar to the FICO score.
• Total balances and debt (11%): This factor evaluates the borrower’s
total debt, including loans and credit card balances.
• Recent credit behavior and inquiries (5%): This factor evaluates recent
credit activity, including the number of new credit accounts and credit
inquiries.
• Available credit (3%): This factor evaluates the borrower’s available
credit, or the amount of credit they could access if they needed it.
•

As technology continues to advance, credit scoring models are evolving to keep

pace with the changing landscape. Here are some emerging trends and
technologies in credit scoring to watch out for:
 1. Big Data:
The use of big data and machine learning algorithms can help lenders analyze
vast amounts of data to identify patterns and make more informed lending
decisions. According to a study, 76% of lenders are already using machine
learning in some capacity to evaluate creditworthiness.
2. Alternative Data:
The use of alternative data sources, such as utility bill payments and rental
history, is becoming more prevalent in credit scoring models. This can help
lenders evaluate borrowers who may not have a traditional credit history and
improve access to credit for underserved populations.
3. Real-Time Scoring:
Real-time credit scoring can provide lenders with up-to-date information on a
borrower’s creditworthiness, allowing for more accurate and timely lending
decisions. This can be particularly useful for small business owners who need
access to credit in a timely manner.
4. Mobile Scoring:
With the rise of mobile banking, lenders are exploring the use of mobile data to
evaluate creditworthiness. This includes analyzing a borrower’s mobile phone
usage patterns, such as the frequency of calls and text messages.
5. Financial Health Scoring:
Financial health scoring models are emerging as a way to provide a more
holistic view of a borrower’s financial health. These models take into account
factors such as savings, investments, and debt levels to provide a more
comprehensive picture of a borrower’s creditworthiness.
Overall, the future of credit scoring models is exciting and full of potential. As
new technologies and trends emerge, lenders and borrowers alike can expect to
see more innovative and effective ways to evaluate creditworthiness and
improve access to credit.
 MARKETING ANALYTICS
Marketing analytics is the process of tracking and analyzing data from
marketing efforts, often to reach a quantitative goal. Insights gleaned from
marketing analytics can enable organizations to improve their customer
experiences, increase the return on investment (ROI) of marketing efforts.
(Cote, 2021)

Where Does Marketing Data Come

The data you use to track progress toward goals, gain customer insights, and
drive strategic decisions must first be collected, aggregated, and organized.
There are three types of customer data: first-party, second-party, and third-party.
• First-party data is collected directly from your users by your organization.
It’s considered the most valuable data type because you receive
information about how your audience behaves, thinks, and feels.
• Second-party data is data that’s shared by another organization about its
customers (or its first-party data). It can be useful if your audience types
 are the same or have similar demographics, if your companies are running
a promotion together, or if you have a partnership.
• Third-party data is data that’s been collected and rented or sold by
organizations that don’t have a connection to your company or users.
Although it’s gathered in large volumes and can provide information
about users similar to yours, third-party data isn’t the most reliable
because it doesn’t come from your customers or a trusted second-party
source.
While it’s important to know that second- and third-party sources exist, first-
party data is the most reliable of the three because it comes directly from your
customers and speaks to their behaviors, beliefs, and feelings. Here are some
ways to collect first-party data.

Surveys
Surveying your current and potential customers is a straightforward way to ask
them about their experiences with your product, their reason for purchasing,
what could be improved, and if they’d recommend your product to someone
else—the possibilities are endless. Surveys can be anything from multi-question
interviews to a popup asking the user to rate their experience on your website.
A/B Tests
An A/B test is a way of testing a hypothesis by comparing user interactions with
a changed version of your website or product to an unchanged version. For
instance, if you hypothesize that users would be more likely to click a button on
your site if it were blue instead of red, you could set up an A/B test in which
half of your users see a red button (the control group) and half see a blue button
(the test group). The data collected from the two groups’ interactions would
show if your hypothesis was correct. A/B tests can be a great way to test ideas
and gather behavioral data.
Organic Content Interaction
Interaction with organic content—such as blog posts, downloadable offers,
emails, social media posts, podcasts, and videos—can be tracked and leveraged
to understand a user’s purchasing motivation, their stage in the marketing
funnel, and what types of content they’re interested in.
Paid Advertisement Interaction
You can also track when someone engages with a digital ad you’ve paid to
display, whether it’s on another website, at the top of search results, or
sponsoring another brand’s content. This data is crucial in determining where
your customers are coming from and what stage of the funnel they see your ads.

PRODUCT MANAGEMENT

Product management analytics is the practice of collecting, interpreting, and

acting on product, user, and business data. The best product analytics go beyond
basic usage metrics. It’s not just about tracking page views or daily active users
— it’s about understanding your users and growth drivers to build a more
successful product.
Traditional analytics tools often trap product data in silos, keeping it separate
from marketing, sales, or support info and making it hard to see the big picture.
(Kazakoff, 1999)

Key use cases for product management

1. Optimizing the user journey to drive product adoption

PMs need to align their teams around creating a product experience users love.
Analytics help you pinpoint areas of frustration across the user journey you can
focus on improving. They can also show you what’s working well at turning
new users into activated, engaged adopters who derive long-term value from
your product.
This might involve analyzing drop-off points in your onboarding funnel, using
qualitative user feedback to understand their journey firsthand, and identifying
key “aha!” moments when users first experience the benefits of your product.
2. Prioritizing features on your roadmap
Analytics data helps PMs make decisions on competing priorities by focusing
on what matters most — to users and the bottom line.
For example, let’s say you’re looking to prioritize feature requests made by
users. Using warehouse-first product management analytics, you could bring
together requests for future features with current feature engagement data and
then segment users to understand how your highest-value customers are using
your main features and what they most need.
Maybe you discover that features requested by enterprise customers, while
fewer in number, have 3x the impact on annual contract value — you may want
to focus initial efforts on those.
Effective product management analytics let you identify the opportunities with
the most potential impact on your key metrics, whether that’s driving adoption,
increasing retention, or boosting revenue.
3. Promoting retention and preventing churn
Understanding what keeps users engaged and what prompts them to leave is
mission-critical for product managers—especially since retaining existing
customers is generally much more cost-effective than acquiring new ones.

Analytics can show you which user behavior patterns correlate with long-term
retention or churn risk.

For example, you might find that users who haven’t logged in to your app for
two weeks are highly likely to churn. Based on this insight, you could trigger a
re-engagement campaign, maybe offering these users a personalized tutorial on
a high-value feature they haven’t yet explored. Or you could focus on slicing
and dicing the data to understand possible reasons why these user groups aren’t
engaged.

Many PMs find it useful to create user “health” scores based on metrics like
number of logins, feature usage, and time active. You can then set up automated
alerts when scores drop below certain thresholds to take early action.

Remember, the goal isn’t just to react to churn when it happens, but to create
such a compelling user experience that users don’t want to leave in the first
place. Product management analytics help you to continuously refine your
product to increase stickiness, meet user needs, and boost long-term
engagement.

4. Iterating and innovating

Continuous improvement is at the heart of effective product management.
Analytics provide the data-driven insights you need to iterate on your product
with confidence. This lets your teams test hypotheses, measure the impact of
changes, and innovate based on real user behavior.

For example, a product team might use A/B testing to test different designs and
functionalities when rolling out a new feature, comparing adoption rates, usage
frequency, and impact on key performance indicators.

5. Forecasting and planning for scalability

Great product managers think ahead. Analytics can help you forecast growth,
predict the resources you’ll need, and plan as you scale.

Using historical user data and predictive analytics, you can anticipate future
trends and prepare your product and infrastructure accordingly. This might
involve analyzing user growth rates, usage patterns, and performance metrics
and even looking at projected revenue patterns for different cohorts.
 MARKETING MIX ALLOCATIONS
A marketing mix includes multiple areas of focus as part of a
comprehensive marketing plan. The term often refers to a common
classification that began as the four Ps: product, price, placement, and
promotion.

Effective marketing touches on a broad range of areas as opposed to fixating on

one message. Doing so helps reach a wider audience, and by keeping the four Ps
in mind, marketing professionals are better able to maintain focus on the things
that really matter. Focusing on a marketing mix helps organizations make
strategic decisions when launching new products or revising existing products.
(Kenton, 2023)

The four Ps classification for developing an effective marketing strategy was

first introduced in 1960 by marketing professor and author E. Jerome
McCarthy.E. Jerome McCarthy. "Basic Marketing: A Managerial Approach,"
Page vi. R.D. Irwin, 1960.

It was published in the book entitled Basic Marketing: A Managerial

Approach. Depending on the industry and the target of the marketing plan,
marketing managers may take various approaches to each of the four Ps. Each
element can be examined independently, but in practice, they often are
dependent on one another.

Product
This represents an item or service designed to satisfy customer needs and wants.
To effectively market a product or service, it's important to identify what
differentiates it from competing products or services. It's also important to
determine if other products or services can be marketed in conjunction with it.

Price

The sale price of the product reflects what consumers are willing to pay for it.
Marketing professionals need to consider costs related to research and
development, manufacturing, marketing, and distribution—otherwise known
as cost-based pricing. Pricing based primarily on consumers' perceived quality
or value is known as value-based pricing.

Value-based pricing plays a key role in products that are considered to be status
symbols.

Placement

When determining areas of distribution, it's important to consider the type of

product sold. Basic consumer products, such as paper goods, often are readily
available in many stores. Premium consumer products, however, typically are
available only in select stores.

Promotion

Joint marketing campaigns are called a promotional mix. Activities might

include advertising, sales promotion, personal selling, and public relations. One
key consideration is the budget assigned to the marketing mix. Marketing
professionals carefully construct a message that often incorporates details from
the other three Ps when trying to reach their target audience. Determination of
the best mediums to communicate the message and decisions about the
frequency of the communication also are important.
 DIGITAL MARKETING

Digital marketing involves measuring, collecting, and analyzing data from

various digital channels. Doing so provides insights into user behavior and how
they interact with digital content.

Relevant data can come from a wide range of sources. This might include
websites, social media, search engines, and email, to name just a few—any
digital channel can provide data that will be useful in some way.

The specific metrics you might analyze are even more diverse, including
everything from website traffic to landing page performance, conversion rates,
engagement rates, and more (see the next section for details).

Whether insights come from dedicated data analysts or digital marketers with an
analytical slant, they can inform real-time strategy on practically any channel.
By supplementing natural intuition and creativity with hard facts about what
works, what doesn’t, and where best to allocate budget and resources, digital
marketing analytics has quickly become an indispensable tool for improving
marketing strategies on the go.

Digital marketing is about more than just crunching numbers. It is an

invaluable tool for understanding your audience, your tactics, and assessing the
effectiveness of your digital marketing spend and overall strategy.
You can track customer behavior and preferences

By tracking customer behavior in real time you can quickly identify content that
resonates best with your audience. Learning their passions and pain points helps
you quickly update content, hitting your audience with the right messages.

You can assess campaign performance in granular detail

Once a campaign has taken place (and even during) digital marketing analytics
helps you measure its effectiveness. Going far beyond merely ‘successful’ or
‘unsuccessful’ you can drill down into specific elements of campaigns, such as
the channel, the timing, or the targeting. You can then make adjustments—even
to ‘successful’ campaigns—to further improve given aspects in the future.
Digital marketing analytics shapes effective online strategies for businesses. It
involves the real-time collection and analysis of data. Advanced tools are then
used to measure online campaign performance, track user engagement, and
evaluate marketing strategy effectiveness. The real action then comes via
interpretation of the data. This is where businesses gain the actionable insights
that help refine their marketing approaches, target specific audience segments,
and enhance overall customer experiences.

When aligned with a strategic approach, the digital marketing analytics process
becomes a powerful tool for businesses to boost their online presence. Here's a
simplified breakdown of the process:
References
Ali, N. (2024, june 11). Lifetime Value Calculation. Lifetime value calculation overview.

Bansal, S. (2020, December 2). Business Forecasting. What is Business Forecasting And Its Methods?

Cote, C. (2021, january 21). Marketing analytics.

jhonson, B. (2023, july 3). credit scoring models. Credit scoring models for better financial future.

Kazakoff, M. (1999, february 03). marketing analytics.

Kenton, W. (2023). Marketing mix.

Wainaina, P. (2023, October 24). Time series forecasting. A complete guide to Time series forecasting.