Analysis of Laptop Price Influencing Factors and Price Prediction

Sihan Guo

and Jihao He

School of Wu Xiang Middle School, Ningbo, 31511, China

School of Mathematics and Physics, Xi’an Jiaotong-Liverpool University, Suzhou,16302, China

Keywords: Laptop Price, Regression Model, Price Prediction

Abstract: This study extensively investigates the intricate factors influencing laptop prices, aiming to demystify the

laptop market's complexities and develop a robust predictive model for pricing. Employing a dataset

comprising 1303 diverse laptop models, the research meticulously examines various attributes such as brand,

screen size, CPU, RAM, storage type, graphics card, operating system, and weight. The study’s thorough data

preprocessing, ensuring data integrity and suitability for analysis. Initial exploratory data analysis (EDA)

uncovers revealing trends and correlations, notably emphasizing the significant impact of brand and technical

specifications on laptop prices. The research methodically employs descriptive statistics to succinctly

summarize the dataset, alongside utilizing graphical representations to vividly illustrate the distribution and

relationship of key attributes. Furthermore, a detailed correlation analysis illuminates the complex interplay

among various features, elucidating their collective influence on pricing. Conclusively, this study not only

provides a comprehensive analysis of the myriad factors affecting laptop prices but also introduces a

sophisticated tool for price prediction. This significant contribution aids consumers in making informed

decisions and offers valuable insights for market analysis.

1 INTRODUCTION

Laptops, once a symbol of luxury and technological

advancement, have now become ubiquitous in almost

every sphere of modern life. Their role is pivotal

across various sectors, impacting the way we work,

learn, and entertain ourselves (Smith & Jones 2019).

People from all walks of life have expanded their

study, work and life from a specific area with limited

space to an infinite dimension because of laptops

(Thompson, et al. 2021). Understanding the factors

influencing their pricing is not just of academic

interest but has practical implications for a range of

stakeholders, from individual consumers to large

corporations.

The laptop market is characterized by rapid

technological advancements. Each new generation of

laptops brings forth innovations in processing power,

battery life, display quality, and overall system

performance. These advancements, while enhancing

user experience, also impact the cost structure of

laptops (Davis & Chung 2022). The development of

new, more efficient processors, for instance, might

initially increase prices but can also lead to cost

reductions as older technologies become obsolete.

This paper will explore these dynamics in detail,

using industry reports and technological forecasts as

references.

Consumer preferences have a significant impact

on laptop pricing. Thompson et al.'s study showed

that with the growing trend of personalization and

customization, manufacturers are increasingly

tailoring their products to meet diverse consumer

needs (Lee & Kim 2018). This section will analyze

how the demand for specific features, like lightweight

design or high-performance graphics, influences

pricing. Studies by Davis and Chung offer insights

into these consumer behavior patterns and their

implications for pricing strategies (Gupta & Malik

2022).

The global economic landscape plays a critical

role in shaping laptop prices. Fluctuations in currency

exchange rates, changes in labor costs, and variations

in raw material prices are all factors that influence the

final cost of laptops (Fernandez & Liu 2023). This

section will delve into how these macroeconomic

factors, along with international trade policies and

supply chain complexities, contribute to the pricing

strategies of laptop manufacturers.

470

Guo, S. and He, J.

Analysis of Laptop Price Inﬂuencing Factors and Price Prediction.

DOI: 10.5220/0012822300004547

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Data Science and Engineering (ICDSE 2024), pages 470-475

ISBN: 978-989-758-690-3

In an increasingly competitive market,

manufacturers are constantly seeking strategies to

gain an edge (Green & Patel 2020). This section will

examine how competition, brand positioning, and

marketing strategies impact laptop pricing.

Additionally, the growing emphasis on sustainability

and environmental responsibility is influencing

manufacturing processes and material choices, which

in turn affects pricing (Brown & Harris 2021). The

research of Gupta and Malik, Fernandez and Liu,

Green and Patel, and Brown and Harris provided

valuable insights into these trends (Williams &

O'Donnell 2022).

The rise of e-commerce has revolutionized the

laptop market, offering consumers a wider range of

choices and more competitive pricing. This section

will analyze the impact of online retail on laptop

pricing, exploring how the digital marketplace has

altered traditional pricing models. Studies by

Williams and O'Donnell highlighted the role of online

platforms in shaping consumer perceptions and

pricing strategies (Chen & Cheng 2023).

Emerging markets present new opportunities and

challenges for laptop manufacturers. This section will

explore how the growing demand in these markets is

influencing global pricing strategies. Additionally,

future trends such as the integration of artificial

intelligence and the evolution of hybrid work

environments will be discussed for their potential

impact on laptop pricing.

This essay aims to construct a comprehensive and

predictive model for laptop pricing, integrating a

diverse range of factors from technological

innovations to global market dynamics. Such an

analysis is vital for understanding the current laptop

market and anticipating future trends. The insights

gained from this study are intended to assist

stakeholders in making informed decisions in a

rapidly evolving market.

2 METHODOLOGY

2.1 Data Source

This study employs a comprehensive dataset from

Kaggle to analyze and forecast laptop prices. The

dataset owned by Muhammet Varlı includes 1,303

entries, each representing a different laptop model.

2.2 Data Cleaning and Preprocessing

All numerical fields were standardized to ensure

consistency, with units of measurement (e.g., 'kg' for

weight, 'inches' for screen size) removed for

quantitative analysis. Missing Value Treatment:

Missing data was addressed based on the nature of the

missing values. For numerical fields, we used mean

or median imputation, while for categorical fields,

mode imputation or removal of records was

employed, depending on the extent of missing data.

Outliers that were deemed to be errors were removed

or corrected based on the context.

2.3 Data Overview

Table 1 provides descriptive statistics of the laptop

dataset, including counts, mean, standard deviation,

minimum, median, and maximum values for various

attributes such as Laptop ID, Screen Size, and Price

in euros.

Table 1: Selected variables

Metric Mean SD Minimum Median Maximum

Laptop_ID 660.16 381.17 - --

Screen Size(inches) 15.02 1.43 10.1 15.6 18.4

Price_euros 1,123.69 699.01 174.00 977.00 6,099.00

Table 2 showcases sample data from the dataset,

illustrating typical entries for attributes like Laptop

ID, Screen Size, Price, Company, Product, and others,

giving a snapshot of the data's structure and content.

Analysis of Laptop Price Inﬂuencing Factors and Price Prediction

471

Table 2: Some examples of other variables

Metric Exam

le Data Exam

le Data

Laptop_ID 1 2 3 4 5

Price_euros 1339.69 898.94 575 2537.45 1803.6

Company Apple Apple HP Apple Apple

Product MacBook Pro Macbook Air 250 G6 MacBook Pro MacBook Pro

TypeName Ultrabook Ultrabook Notebook Ultrabook Ultrabook

Screen

Resolution

2560x1600 1440x900 1920x1080 2880x1800 2560x1600

CPU(GHz) i5 2.3 i5 1.8 7200U 2.5 2.7 3.1

RAM(GB) 8 8 8 16 8

Memor

(

)

128 SSD 128 Flash Stora

e 256 SSD 512 SSD 256 SSD

GPU

Intel Iris Plus

Gra

hics 640

Intel HD Graphics

6000

Intel HD

Gra

hics 620

AMD Radeon

Pro 455

Intel Iris Plus

Gra

hics 650

OpSys macOS macOS No OS macOS macOS

Weight(kg) 1.37 1.34 1.86 1.83 1.37

2.4 Exploratory Data Analysis (EDA)

This included calculating mean, median, standard

deviation, and range for numerical variables to

understand the data distribution. Frequency

distributions for categorical variables were analyzed,

and histograms and box plots were used for numerical

variables to visualize data distribution. Correlation

coefficients were calculated to identify potential

linear relationships between numerical variables,

especially between specifications and price.

2.5 Predictive Modeling

Based on EDA insights and correlation analysis,

features that significantly influenced laptop prices

were selected for model building. In this paper, we

will build a linear regression model. The model's

choice was based on performance metrics such as R-

squared, Mean Absolute Error (MAE), and Root

Mean Squared Error (RMSE). The final model was

chosen based on its performance across multiple

evaluation metrics.

3 RESULTS AND DISCUSSION

3.1 Descriptive Statistics and

Distribution Analysis

The dataset's exploration revealed an average laptop

price of €1123.69, with a range extending from €174

to €6099. The screen sizes varied from 10.1 to 18.4

inches, averaging at 15.02 inches. Brand distribution

analysis highlighted a significant market presence of

manufacturers like Dell, Lenovo, and HP.

3.2 Exploratory Data Analysis

This table 3 provides information on the sample size

used in the analysis, detailing the count of valid and

invalid samples, ensuring clarity on the dataset's

integrity and the extent of data used.

Table 3: Summary of missing samples

Item Sample size

Proportion of

total

Valid sample 1303 100.0%

Exclude invalid sample 0 0.0%

total 1303 100%

Figure 1 displays the distribution of laptop prices.

It shows the range and common price points of

laptops in the dataset.

Figure 1: Distribution of Laptop Prices (Picture credit:

Original).

ICDSE 2024 - International Conference on Data Science and Engineering

472

Figure 2: Average Laptop Price by Brand (Picture credit:

Original)

Bar chart of Figure 2 illustrates the average price

of laptops for each brand, highlighting brands with

higher or lower average prices.

From the table 4 shows that will laptop_ID,

Company, Cpu, ScreenResolution, TypeName,

Inches, Memory, Ram, Weight, OpSys, Gpu as

independent variables, Price euros is used as the

dependent variable for linear regression analysis. As

can be seen from the above table, the formula of the

model is as follows:

𝑃𝑟𝑖𝑐𝑒



= −1884.483 + 0.15 × 𝑙𝑎𝑝𝑡𝑜𝑝



8.171 × 𝐶𝑜𝑚𝑝𝑎𝑛𝑦 + ⋯ + 6.887 × 𝐶𝑝𝑢

(1)

Table 4: Linear regression analysis results

Standard

error

Beta t p VIF tolerance

constant -1884.483 272.212 - -6.923 0.000** - -

laptop_ID 0.150 0.038 0.082 4.005 0.000** 1.043 0.958

Company 8.171 3.564 0.048 2.293 0.022* 1.078 0.927

Cpu 9.603 0.657 0.346 14.625 0.000** 1.393 0.718

ScreenResolution 18.290 1.969 0.206 9.291 0.000** 1.226 0.815

TypeName 0.219 12.452 0.000 0.018 0.986 1.224 0.817

Inches 75.761 20.745 0.155 3.652 0.000** 4.456 0.224

Memory 10.799 1.882 0.124 5.739 0.000** 1.156 0.865

Ram -46.121 5.792 -0.174 -7.964 0.000** 1.181 0.847

Weight -2.357 0.713 -0.145 -3.303 0.001** 4.796 0.208

OpSys 127.711 13.960 0.189 9.148 0.000** 1.063 0.941

Gpu 6.887 0.815 0.203 8.448 0.000** 1.432 0.698

R 2 0.481

Adjusted R 2 0.476

F F (11,1291) =108.726, p=0.000

D-W-value 2.056

* p<0.05 ** p<0.01

In the linear regression analysis of laptop pricing,

several key components emerged as significant

influencers. The CPU and GPU, with coefficients of

9.603 and 6.887 respectively, are substantial positive

drivers of price, reflecting the importance of

processing and graphics capabilities in determining

laptop value. Screen resolution, represented by a high

coefficient of 18.290, also positively influences the

price, indicating a market preference for higher-

quality displays. Interestingly, the operating system

(OpSys) showed a remarkably high coefficient of

127.711, suggesting its pivotal role in laptop pricing.

The size of the laptop, as denoted by 'Inches' with a

coefficient of 75.761, positively impacts the price,

aligning with the trend towards larger, more

expensive models (Table 4).

Conversely, certain features exhibited a negative

relationship with pricing. Notably, RAM, with a

coefficient of -46.121, surprisingly indicates a

decrease in price with increased RAM, a finding that

contradicts typical market expectations and may

warrant further investigation. Similarly, the weight of

the laptop negatively affects its price (coefficient -

2.357), hinting at a consumer preference for lighter,

more portable models.

Analysis of Laptop Price Inﬂuencing Factors and Price Prediction

473

The brand (Company) and storage capacity

(Memory) also play significant roles, with

coefficients of 8.171 and 10.799 respectively,

underlining the influence of brand reputation and

storage on pricing. Interestingly, the laptop's unique

identifier (Laptop_ID) showed a small but significant

positive effect (coefficient 0.150), possibly reflecting

a trend towards higher pricing in newer models. In

contrast, the laptop type (TypeName), despite having

a coefficient of 0.219, did not significantly impact the

price, suggesting a lesser role for this feature in

determining laptop value.

3.3 Predictive Model Performance

From the table above shows that will laptop_ID,

Company, Cpu, ScreenResolution, TypeName,

Inches, Memory, Ram, Weight, OpSys, Gpu as

independent variables. For linear regression analysis

using Price_euros as the dependent variable, it can be

seen from the table above that the R-square value of

the model is 0.481. Means laptop_ID, Company, Cpu,

ScreenResolution, TypeName, Inches, Memory,

Ram, Weight, OpSys, Gpu can explain why

Price_euros 48.1% change.

Table 5: Model summary

R R 2

Adjusted

R 2

Model

error

RMSE

DW-

value

AIC-

value

BIC-

value

0.6930.481 0.476 503.433 2.056 19934.857 19996.926

Table 6: ANOVA form.

Sum of squares df Mean square F p-value

Regression 305935781.750 11 27812343.795 108.726 0.000

Residual error 330239179.540 1291 255801.069

Total 636174961.290 1302

As can be seen from the above table 5 and 6, it is

found that the model passes the F test (F=108.726,

p=0.000<0.05) when F-test is performed on the

model, which means that the model construction is

meaningful. Although other models like Decision

Trees and Random Forests were considered, the

Linear Regression model was chosen for its

simplicity and interpretability, which are particularly

advantageous in understanding how different features

impact laptop prices linearly.

3.4 Discussion

The results highlight the complexity of laptop pricing.

The positive influence of CPU and Screen Resolution

suggests that higher performance and better display

quality are valued in the market, leading to higher

prices. The negative impact of Ram is intriguing and

might suggest market trends favoring efficient,

compact laptops over higher memory capacities. The

influence of OpSys indicates a brand-specific

preference in pricing, aligning with consumer

behavior and market segmentation. The moderate R-

squared value points to the presence of other

influencing factors not captured in the model,

suggesting the need for a more comprehensive

approach or the inclusion of additional variables like

brand reputation or specific features like battery life

or build quality. These findings provide valuable

insights for manufacturers and retailers, suggesting a

focus on specific features to align with market

demands. Additionally, they offer guidance to

consumers on what factors to consider when

assessing the value of a laptop.

4 CONCLUSION

The linear regression analysis on laptop prices

revealed the significant impact of various factors like

CPU, Screen Resolution, Inches, Memory, OpSys,

and GPU, each positively influencing the price.

Conversely, variables such as Ram and Weight

demonstrated a negative impact. The model explained

48.1% of the price variance, indicating a moderate

level of prediction accuracy. The study underscores

the complexity of laptop pricing, influenced by a

blend of technical specifications and brand-related

factors. The significant positive impact of CPU and

Screen Resolution aligns with the market's emphasis

ICDSE 2024 - International Conference on Data Science and Engineering

474

on performance and display quality. The negative

influence of Ram suggests a nuanced market

preference, potentially favoring portability over high

memory capacity. The notable effect of operating

systems highlights brand-specific preferences in

laptop pricing. These insights are invaluable for

manufacturers and retailers, guiding them to align

their products with consumer preferences and market

trends. For consumers, understanding these factors

can aid in making informed purchasing decisions.

Future research could enhance model accuracy by

including additional variables such as brand

reputation, specific technical features, and evolving

market trends. This study contributes to a deeper

understanding of the laptop market, offering a

foundation for further research and practical

application in pricing strategies and consumer

behavior analysis.

AUTHORS CONTRIBUTION

All the authors contributed equally and their names

were listed in alphabetical order.

REFERENCES

A. Fernandez and C. Liu, Emerging Markets Review 30(2),

150-165 (2023).

A. Smith and B. Jones, Journal of Technology and

Economics 34(2), 102-117 (2019).

J. Lee and Y. Kim, Global Economy Journal 22(4), 337-

355 (2018).

K. Brown and J. Harris, Journal of Policy and Economics

17(2), 233-247 (2021).

M. Williams and L. O'Donnell, Journal of Online Retailing

5(3), 200-216 (2022).

R. Davis and H. Chung, Brand Management Journal 19(4),

112-130 (2022).

R. Thompson, et al., Market Dynamics Review 12(3), 200-

215 (2021).

S. Gupta, P. Malik, Journal of Business Strategies 29(1),

78-92 (2022).

T. Green and S. Patel, Journal of Sustainable Production

18(5), 275-289 (2020).

Z. Y. Chen and L. Y. Cheng, Computer Science and

Applications 13(8), 1538-1546 (2023).

Analysis of Laptop Price Inﬂuencing Factors and Price Prediction

475