IB Subject: Mathematics AI

Extended Essay

How can statistical regression and trend analysis quantify the

disruption caused by e-commerce to traditional retail

businesses?

Word Count: 3412

Table of Contents
Introduction

E-Commerce and Its Impact

Historical Sales Data Analysis
Changing Consumer Behavior

Mathematical Methods
Regression Models
Trend Analysis
ARIMA Forecasting

Data Analysis
Market Share Trends
Sectoral Disruptions
Model Validation

Discussion
Strategic Implications
Limitations and Ethical Considerations

Conclusion

References
Appendices

Introduction:

2
The international economy has drastically changed as a result of the e-commerce industry's

explosive growth in recent years. Online platforms' ease, accessibility, and broad product selection

have significantly changed customer expectations and behaviours, resulting in a fundamental shift in

the way that goods and services are bought. Traditional brick-and-mortar retail companies have had

to reevaluate their business plans and tactics in order to stay competitive as a result of the

development of e-commerce. The economic and societal variables that formerly supported physical

businesses have changed as more customers choose the ease of internet buying. In addition to

having an effect on retailers' bottom lines, this development has had a major influence on market

share, customer preferences, and general industry trends.With the use of statistical regression and

trend analysis, this Extended Essay seeks to investigate and measure the disruption that

e-commerce has caused to conventional retail businesses. In the context of this digital transition,

these mathematical methods will be employed to examine how sales trends, changes in market

share, and changing customer behaviour may be quantified and comprehended. The essay will use

regression approaches and trend analysis to shine the spotlight on how much e-commerce has

changed the retail industry, how traditional retailers have responded to these changes, and whether

statistical techniques may predict how the industry will develop in the future in an increasingly digital

world. Moreover, this essay will also show how these models may help conventional shops in

identifying areas for adaptation, directing strategic choices to maintain competitiveness in an

economy that prioritises digitalisation. This study will close the gap between theory and practice by

combining mathematical methods with empirical data, offering practical insights into one of the

biggest shifts in contemporary trade.

Background:

3
In order to respond to the enquiries,“How can statistical regression and trend analysis quantify the
disruption caused by e-commerce to traditional retail businesses?”. This essay will use an organised
mathematical method backed by relevant data. In order to create accessible links between the
growing popularity of e-commerce and its effects on conventional retail sales, the study will
concentrate on gathering and evaluating numerical information from trustworthy sources.

Sources of Data

1.​ Historical Sales Data: To find patterns and changes in market share, annual sales data for
traditional retail companies and e-commerce platforms will be gathered for a minimum of ten
years. The sources of these datasets will be:
○​ Market research reports (e.g., Forrester Research, Statista, Nielsen).
○​ Government databases (e.g., U.S. Census Bureau, Eurostat).
○​ Published financial statements from major retail corporations.
○​ Etc (specified down below in sources)

2. Consumer Behavior Studies: Additional context for comprehending changes in consumer

behaviour will be provided by statistical information on consumer preferences, frequency of
shopping, and spending patterns that is obtained from surveys or industry reports.

3. Economic Indicators: More broad economic metrics, such as GDP growth rates and inflation, will
be considered to account for external factors influencing retail trends.

4. Case Studies: Selected case studies of traditional retailers (e.g., Macy’s, Debenhams) and
e-commerce giants (e.g., Amazon, Alibaba) will be analyzed to make more sense the mathematical
findings.

4
 Methodology

Several mathematical tools will be used to quantify the disruption that e-commerce has caused to

traditional retail. These techniques will allow for future predictive forecasting and offer a thorough

examination of the relationship between the rise in e-commerce sales and the fall in traditional retail.

The investigation's primary methods and strategies are described in the sections that follow:

1. Statistical Regression Analysis: Objective: To measure the correlation between traditional

retail sales declines and e-commerce sales growth.

To determine the link between the two variables of interest—e-commerce sales and traditional retail

sales—regression models will be used. Understanding the degree of disruption brought about by the

expansion of e-commerce will be made possible thanks in large part to these models. The following

important variables will be part of the regression analysis:

-E-commerce sales (as a percentage of total retail sales).

-Traditional retail sales (broken down by sector, such as clothing, electronics, and food).

-Consumer spending patterns, which reflect shifts in consumer behavior.

-Demographic factors (e.g., age, income level, geographical location) that may influence the

adoption of e-commerce over traditional retail.

I will look into both linear and multivariate regression approaches. While ARIMA takes into account

the time-dependent nature of sales data and may capture seasonal effects, these more

straightforward regressions provide a baseline for comparing results. Because of this, ARIMA may

be capable of predicting monthly or quarterly retail patterns and measure the disruption brought on

by e-commerce with more accuracy. One of the main strategies to be used is ARIMA

(Auto-Regressive Integrated Moving Average)..

5
The direct correlation between the increase in e-commerce sales and the decrease in traditional

retail sales will first be investigated using a linear regression model. For instance, the linear model

will capture the link between a 0.5% drop in traditional retail sales and a 1% increase in e-commerce

sales.

2. Trend Analysis:

Objective: is to examine past sales data in order to spot cyclical shifts and long-term growth trends in

the traditional retail and e-commerce industries.

In order to find both short-term and long-term patterns, time series analysis will be used to look at

historical data on traditional retail sales and e-commerce. In order to smooth out data outliers and

draw attention to the deeper trends in consumer behaviour, average moving and exponential

smoothing techniques will be used. These techniques will be used to compare the growth paths of

traditional retail sales data and e-commerce data.

Exponential smoothing is specifically selected because it may provide more weight to recent

observations, which makes it more responsive to recent changes in the market. In order to smooth

out any seasonal variations that could exist in the retail and e-commerce industries, moving

averages will also be utilised to track the general trend over time.

3. Forecasting Models:

Objective:to develop predictive models that project future developments in traditional retail sales as

well as e-commerce

To predict baseline forecasts using more straightforward models, such as linear or multiple

regression, in order to estimate future patterns in e-commerce and conventional retail sales.

However, an important approach to forecasting time-series data is then presented: ARIMA

(Auto-Regressive Integrated Moving Average). Because ARIMA can capture both trend and

seasonal components that are typical in retail data, including Christmas surges or back-to-school

purchasing periods, it was selected above simpler models. This ability is essential for correctly

6
predicting sales and understanding how e-commerce affects different times of the year. The ARIMA

model is quite useful as it can handle both trends and seasonality in data. By comparing expected

values with actual historical data, forecasting models will be evaluated.I can show if improved

accuracy in predictions and a more clear explanation of the observed patterns justify the complexity

of ARIMA by comparing its performance with that of linear models.

Confidence intervals and mean squared error will be used to evaluate how accurate the forecasts

were.

7
 Data Analysis Framework

The analysis will proceed in the following steps:

1.​ Data Collection and Cleaning:

○​ Mix information from several sources, making sure it is reliable and consistent.

○​ Apply the proper statistical techniques to outliers or missing values.

2.​ Exploratory Data Analysis (EDA):

○​ Use trends using graphs, scatterplots, and histograms.

○​ Determine the statistics that are descriptive (mean, median, variance) to summarize

the data.

3.​ Model Application:

○​ To figure out the degree of e-commerce problems, use regression analysis.

○​ Use trend analysis to assess the historical evolution of retail sales.

4.​ Validation:

○​ Evaluate the reliability of models using statistical metrics such as R-squared,

p-values, and confidence intervals.

○​ Compare model predictions with actual data to assess accuracy.

Challenges and Limitations

1.​ Data Availability: Access to consistent and detailed historical data may be limited,

particularly for smaller retailers or emerging markets.

2.​ External Variables: Factors such as economic recessions, global pandemics, or

technological advancements could skew the results, making it challenging to isolate the

impact of e-commerce.

3.​ Model Assumptions: The validity of regression and forecasting models depends on the

assumptions made during their construction. Deviations from these assumptions could affect

the reliability of the results.

8
 Ethical Considerations

The inquiry will guarantee that all information utilised comes from ethical and publicly accessible

sources. To prevent bias in the interpretation of outcomes, care will be taken to present findings

objectively.

Knowledge and Understanding of the Topic

This section shows a thorough and complex comprehension of how e-commerce's impact on

traditional retail firms may be measured using mathematical methods like statistical regression and

trend analysis. In order to create an evidence-based narrative that answers the research question,

the emphasis is on integrating quantitative methods with actual facts.

Application of Statistical Regression:

Statistical regression models are employed to establish the relationship between e-commerce

growth and the decline in traditional retail sales.

For example:

●​ Linear Regression: Determines the relationship between the percentage increase of

e-commerce sales relative to retail sales and time (as an independent variable).

●​ Multiple Regression: Add additional variables such as population density, income levels,

and internet penetration rates to refine predictions.

The mathematical methods used to analyse and measure the damage that e-commerce has brought

to traditional retail enterprises are fundamental to the topic's comprehension. This criteria combines

theoretical ideas with real-world applications to show in-depth knowledge:

9
Data analysis:

Examining e-commerce's increasing share of overall retail sales is one of the most straightforward

methods to gauge the disruption it has brought about to traditional retail. The graph of Statista's

Global E-commerce Market Share (2021–2027) clearly shows this important change. E-commerce

made up 18.8% of all retail sales worldwide in 2021. This percentage is anticipated to rise gradually

over the next years, reaching 22.6% by 2027. This steady development pattern reveals a distinct and

undeniable trend: e-commerce is steadily gaining market share, directly competing with and

occasionally displacing traditional brick-and-mortar retail establishments.

10
The steady rise in e-commerce's market share may be attributed to many important factors that have

radically changed consumer behaviour and retail actions and deals. One of the main factors is the

notable progress is in logistics. Companies like Amazon, which provide quicker delivery choices that

are frequently more convenient than going to physical stores, have completely changed the pace

and efficiency of order fulfilment. The logistical capabilities of e-commerce platforms have greatly

increased due to the establishment of fulfilment centres strategically situated close to major urban

areas and the growth of distribution networks, which has made online buying more alluring.

The increasing trust that consumers have in online buying platforms is another contributing element.

Consumer reluctance to make purchases online, including worries about security and product

11
legitimacy, has decreased over time. Customers are now more at ease with the concept of making

purchases online because to the development of safe payment methods, thorough product

evaluations, and enhanced return guidelines. A general increase in digital literacy has hastened this

change in consumer behaviour, making internet buying more accessible to a wider spectrum of

individuals, including older generations and tech-savvy millennials.

To measure the rate of expansion of e-commerce, a linear or exponential regression model

may be created using the data points in the Statista graph (2021–2027). A substantial

negative correlation that shows the negative link between the two may be seen when

contrasting this with a decrease in conventional retail sales. The Statista graph's steady rise

highlights how widely e-commerce is being used. The increase from 18.8% in 2021 to 21.8%

in 2026, for example, represents a three percentage point net gain over five years, suggesting

steady, predictable growth trends.

Criterion E: Reasoned Arguments

The idea that traditional retail would be disrupted by e-commerce is directly supported by the graph.

Consumer spending shifts of billions of dollars correspond to every percentage point growth in

e-commerce's retail share. The need for established businesses to take on hybrid models that

combine online and physical channels is highlighted by this measurable effect.

To strengthen the overall analysis in Criterion E and offer deeper insights, the graph that shows the

increase of e-commerce market share should be improved. Although the graph successfully

illustrates the consistent growth of e-commerce from 18.8% in 2021 to 21.8% in 2026, it is now

12
devoid of several components that would greatly enhance its interpretive value and capacity to

bolster the essay's thesis.

First off, the graph might be improved by adding information on both the rise of e-commerce and

traditional retail sales. The negative relationship between these two trends—that is, the idea that

conventional retail declines as e-commerce grows—would be made graphically clear by contrasting

them. By clearly connecting the growth of e-commerce to the disruption in traditional retail, this dual

depiction would not only support the thesis but also make it more powerful and trustworthy.

In addition, the nature of e-commerce's growth would be highlighted by fitting an exponential

trendline to the growth data. The growth of e-commerce shows an exponential trend, pushed by

elements including raised logistics, shifts in customer behaviour, and technology improvements.

Showing the trendline and its equation will show the statistical basis of the argument and offer

concrete evidence of this growth pattern.

Alongside these modifications, sector-specific statistics, such as the effect of e-commerce on the

food, fashion, and electronics industries, might be involved in the graph. A more detailed knowledge

of the ways in which many industries are impacted by the growth of e-commerce would be made

possible by that level of detail. For instance, grocery and food retail may show slower rates of

disruption because of logistical issues, whereas fashion and electronics have mostly moved to online

platforms. Such observations would improve the discourse and fortify the essay's depth of

evaluation.

For the reason that to provide historical and economic context, the graph should also include

contextual indicators. Important periods, like the COVID-19 pandemic or important technical

advancements, might be noted to demonstrate how they sped up the adoption of e-commerce. In

addition to improving the graph's ease of reading, these markers would provide an easier to

understand connection between it and the larger economic patterns covered in the article.

Criterion F: Application of Analytical and Evaluative Skills

13
Applying analytical and evaluative abilities is essential to proving that you have a thorough

comprehension of the study issue. These abilities will be demonstrated in this study by the

methodical examination of quantitative data, the critical assessment of patterns, and the synthesis of

results to answer the research question.

Statistical regression models are first used in the investigation to determine the connections between

the expansion of e-commerce and the decline of traditional retail. The study will quantify connections

using techniques like linear and exponential regression, allowing for a data-driven assessment of the

ways in which e-commerce affects several conventional retail sectors. For instance, in order to find

trends and their statistical significance, variables including consumer spending, e-commerce sales,

and demographic characteristics will be carefully analysed. To guarantee the validity and applicability

of the results, the regression outputs—coefficients, R-squared values, and p-values—will be

analysed.

Analysing the sectoral differences in the effects of e-commerce is another crucial component of

assessment. Through the examination of data from a range of retail businesses, including fashion,

electronics, and groceries, the study will investigate the ways in which the emergence of online

platforms impacts various sectors. The overall case will be strengthened by the nuanced viewpoint

provided by this sectoral research, which will points out differences in change rates. The limits of the

investigation will be acknowledged and addressed in order to guarantee a critical and reflective

attitude. Issues including the availability of data, model assumptions, and outside influences will all

be carefully considered to see how they could affect how reliable the results are. To guarantee

honesty and accuracy in the study, for instance, any departures from regression models'

presumptions would be discussed. Finally, a thorough and unbiased analysis of the research issue

will be possible thanks to the analytical and evaluative abilities used in this study. The essay shows a

high level of critical thinking and intellectual dedication by carefully analysing evidence, synthesising

conclusions, and addressing limits. This method guarantees that the findings obtained are not only

supported by solid data but also accurately represent the complex nature of the subject.This study

has wider implications for understanding the dynamics of e-commerce and its revolutionary

consequences on traditional retail markets, in addition to answering the main research topic. The

14
investigation's findings have practical implications for industry stakeholders in addition to being

important from an academic one. To ensure that they can survive in an increasingly digital economy,

sellers, for example, might utilise the results to modify their strategy by implementing hybrid business

models that connect online and physical retail channels. These findings might also help

policymakers create well-informed rules that strike a balance between the expansion of e-commerce

and the long-term viability of regional companies and jobs.In summary, this essay's analytical and

evaluative strategies show a thorough and careful consideration of the study subject. The research

makes sure that its conclusions are reliable and representative of the detailed reality of the retail

sector by combining solid quantitative data, evaluating trends critically, and solving fundamental

limits. In addition to demonstrating the significant influence of e-commerce on conventional retail,

this methodical approach underlines the importance of thorough analysis in navigating a time of swift

economic and technical change.In summary, this essay's analytical and evaluative strategies show a

thorough and careful consideration of the study subject. The research makes sure that its

conclusions are reliable and representative of the detailed reality of the retail sector by combining

solid quantitative data, evaluating trends critically, and solving fundamental limits. In addition to

showing the significant influence of e-commerce on traditional retail, this systematic strategy

underlines the importance of thorough analysis in navigating a time of swift economic and technical

change.

Criterion G: Engagement

By examining both the quantitative and qualitative facets of the e-commerce phenomenon, I have

actively engaged with the research issue throughout the creation of this essay. A genuine interest in

comprehending how digital revolution alters industries and influences consumer behaviour led to the

choice of this topic. My never-ending curiosity led me to study economic theories and statistical

models in great detail and relate them to current retail structures.One of the most important parts of

my investigation was getting over difficulties while conducting the research. For instance, the early

difficulty of having or getting or even finding access to large databases required creative solutions to

problem-solving, such as merging data from many sources and washing data to guarantee accuracy

and consistency. These actions not only helped me get a deeper technical knowledge of data

15
analysis, but they also made me see how crucial commitment is to academic research.I also thought

about the moral issues of my research, like the limitations of my models and any biases in the

interpretation of the results. This understanding impacted on how I presented my findings, making

sure that all assumptions and limitations were openly addressed and that conclusions were

well-supported. For example, I understood how external things that we cannot control can influence,

for example changes in the global economy, could affect outcomes and evaluated the limitations of

regression models.Another crucial component of my process was interacting with peer and advisor

input. I improved my technique, made my arguments more clear, and made my analysis clearer as a

result of their comments and recommendations. I was able to approach the research with an open

mind and keep improving my work because of this iterative procedure. Additionally this topic helped

me understand the market behaviour in real life. It not only answered the question, but it actually

stamped the information in my brain and made a more intellectual connection and implication of

E-Commerce in today's society.My research was thorough, personally meaningful, and relevant to

larger economic and technological concerns because of my curious and reflective approach.

The Extended Essay procedure has greatly improved my comprehension of economics and

mathematics, the two fundamental fields at the heart of my research. Through the use of

mathematical methods like exponential modelling and regression analysis, I gained a greater

understanding of how abstract mathematical ideas may be applied to practical issues. Making and

analysing these models helped me hone my technical abilities while also bringing to light the

accuracy and constraints of mathematical approaches in economic settings. I was able to develop an

advanced understanding of data-driven decision-making, for example, by critically assessing the

strength of correlations between variables through the analysis of regression outputs like R-squared

values and p-values.

16
 Reference List

Alam, M. (2024) 25 popular in-store vs online shopping statistics, Mailmodo. Available at:
https://www.mailmodo.com/guides/in-store-vs-online-shopping-statistics/?

Anderson, M. (2025) Rosy December for small business sales, but restaurants are left out,
AP News. Available at:
https://apnews.com/article/small-business-holiday-sales-retail-09e31a1e52a10e1a25af5c22
ed804ead

Barnett, J. (2025) Falling retail sales highlight fragility of the economy, The Times. Available
at:
https://www.thetimes.com/business-money/economics/article/falling-retail-sales-highlight-fra
gility-of-economy-8zcrbkjvk

17
Ecommerce vs. Retail: The real differences for businesses and customers (2022)
BigCommerce. Available at:
https://www.bigcommerce.com/articles/ecommerce/ecommerce-vs-retail/? (Accessed:
January 28, 2025).

Global e-commerce share of retail sales 2027 (no date) Statista. Available at:
https://www.statista.com/statistics/534123/e-commerce-share-of-retail-sales-worldwide/

Trangle, S. (2025) Holiday sales topped expectations. That could mean a strong start to
2025, Investopedia. Available at:
https://www.investopedia.com/holiday-sales-topped-expectations-that-could-mean-a-strong-
start-to-2025-8775947?

US Census Bureau (2024) “E-commerce.” Available at:

https://www.census.gov/library/publications/time-series/e-commerce.html?utm_source=chat
gpt.com

US E-Commerce Sales as Percent of Retail Sales quarterly trends: Quarterly E-commerce

report (no date) Ycharts.com. Available at:
https://ycharts.com/indicators/us_ecommerce_sales_as_percent_retail_sales? Villa, L.P.
(2025) El comercio electrónico toma impulso en 2024 con un crecimiento del 18%,
Ediciones EL PAÍS S.L. Available at:
https://cincodias.elpais.com/companias/2025-01-21/el-comercio-electronico-toma-impulso-e
n-2024-con-un-crecimiento-del-18.html?

Young, J. (2022) A decade in review: Ecommerce sales vs. total retail sales 2012-2021,
Digital Commerce 360. Digital Commerce 360 | Internet Retailer. Available at:
https://www.digitalcommerce360.com/article/e-commerce-sales-retail-sales-ten-year-review/
?

Cbre.com. Available at:

https://www.cbre.com/insights/articles/omnichannel-what-is-the-share-of-e-commerce-in-ove
rall-retail-sales?

(Euromonitor.com. Available at: https://www.euromonitor.com

18