Analysis of Factors Influencing Smartphone Prices
Jingchen Cui
1
a
, Xinyu Wang
2,
＊
b
and Yichen Zheng
3
c
1
School of Mathematics, Southwest Jiaotong University, Chengdu 610097, China
2
School of Statistics and Mathematics, Beijing Jiaotong University, Beijing, 100044, China
3
School of Statistics and Mathematics, Shandong University of Finance and Economics, Jinan, 250002, China
＊
Keywords: Smartphone Prices, Influencing Factors, Linear Regression.
Abstract: As a necessity of modern life, the price of smartphones is affected by multiple factors, such as hardware cost,
technological innovation, and marketing strategy. To deeply investigate the common role of various factors
on smartphone pricing, this study is based on a dataset containing 980 basic attributes of smartphones,
improves the data quality through preprocessing steps such as dealing with missing values and deleting
outliers, and filters out the significant variables by using backward stepwise regression. On this basis, this
study constructs a multiple linear regression model by linear regression method, conducts regression analysis
and concludes that factors such as processor speed, Random Access Memor (RAM) size and internal storage
capacity have a positive impact on smartphone price, while factors such as user rating, processor cores and
battery capacity have a negative impact. The results of the study not only provide consumers with a scientific
basis for purchasing decision-making but also provide theoretical support for manufacturers to optimise
product pricing strategies and improve market competitiveness, which is of great significance in promoting
the healthy development of the smartphone market.
a
https://orcid.org/0009-0003-3609-6665
b
https://orcid.org/0009-0003-3088-5778
c
https://orcid.org/0009-0005-8294-8966
1 INTRODUCTION
With the advancement of technology and consumer
demand, the smartphone market has shown a
booming trend. Major mobile phone brands continue
to compete for market share through technological
innovation and marketing strategies. Consumers'
attention has gradually shifted to product features,
performance, and cost-effectiveness. (Wang & Wang,
2022). In this context, it is of great practical
significance to study the influencing factors of
smartphone prices in depth. For consumers,
clarifying the influencing factors of price will help
them better understand the differences between
products at different price levels so that they can
make the best choice according to their own needs
and budgets. For enterprises, mastering these key
factors not only helps to develop a scientific pricing
strategy but also provides a strong basis for new
product development and marketing, helping
enterprises to occupy a favourable position in the
fierce market competition and achieve sustainable
profitability (Qin & Ren, 2021).
At present, the price of smartphones is affected by
multiple factors, such as production costs and
technological innovation, and price-related research
is a hot topic in current market research, which has
been studied by many scholars. Ahmed, Ahmad, &
Bashir (2022) analyzed the effect of mobile phone
attributes on price through the Hedonic price model
and found that 4G mobile phones are significantly
more expensive than 3G mobile phones and that
consumers are willing to pay this premium for faster
data transfer speeds and more reliable network
technology. Tanveer et al. (2021) noted that price has
a significant effect on mobile phone purchasing
behaviour among young people, while factors such as
convenience, Liu & Mo (2021) used event-related
potentials (ERPs) to investigate the potential neural
mechanisms of the price of the reviewer's mobile
phone on consumers’ purchase intention, and found
Cui, J., Wang, X. and Zheng, Y.
Analysis of Factors Influencing Smartphone Prices.
DOI: 10.5220/0013814100004708
Paper published under CC license (CC BY-NC-ND 4.0)
In Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy (IAMPA 2025), pages 65-69
ISBN: 978-989-758-774-0
Proceedings Copyright © 2025 by SCITEPRESS – Science and Technology Publications, Lda.
65
that the price of the reviewer's mobile phone had a
positive effect on consumers’ purchase intention.
The academics also looked at how to build more
accurate price models to better predict and analyse
mobile phone prices.
Chen (2024) improved the hedonic price index
model by incorporating LASSO regression and the
RYGEKS index method, effectively addressing
multicollinearity issues among variables. Xu (2022)
adopted multiple machine learning classification
algorithms for mobile phone price categorization and
prediction, identifying logistic regression as the
optimal model. Han, Li, & Du (2022) proposed an
adaptive price adjustment method based on a Dual
Deep Fuzzy Network (DDFN) for the second-hand
mobile phone market, ensuring accuracy and
reliability in recycled device price adjustments.
Existing studies demonstrate that mobile phone
pricing is influenced by complex multidimensional
factors, necessitating the adoption of advanced
modeling approaches to enhance price prediction
accuracy.
This study focuses on the pricing mechanism of
smartphones, exploring in depth the key factors
affecting smartphone pricing by referring to industry
reports and analysing multi-dimensional data.
Aiming at the problems of insufficient model
accuracy and time lag in existing research on price
influencing factors, this study constructs a more
comprehensive linear regression model by analysing
relevant pricing factors. The study aims to provide a
two-way reference for consumers and enterprises and
promote the healthy development of the smartphone
market.
2 LINEAR REGRESSION
METHODS AND DATA
SOURCES AND
PRE-PROCESSING
Linear regression methods are mainly used to study
the linear relationships between variables and to
model them for prediction and data analysis (Maulud
and Abdulazeez, 2020). This study constructs a
multiple linear regression model with smartphone
price as the response variable and each key factor
affecting mobile phone price as the predictor, and
analyses the correlation between these factors and
smartphone price.
This study uses a comprehensive dataset from the
Kaggle website, which presents an all-encompassing
collection of information on all the latest
smartphones existing in the market, which can be
used for in-depth analysis of the factors affecting
smartphone pricing (Kaggle, 2023). The dataset was
created by Abhijit Dahatonde and contains basic
attributes of 980 different types of smartphones,
covering a wide range of information such as brand,
model, configuration, etc (Kaggle, 2023).
Before analyzing the collected dataset, data
preprocessing was conducted to enhance data quality
and ensure analytical reliability. First, missing data
were addressed through mean or median imputation
based on variable distributions. Subsequently, price
values were converted from Indian Rupees (INR) to
US Dollars (USD) using the exchange rate to
standardize numerical variables for quantitative
analysis. Finally, outliers were identified and
addressed through rigorous statistical inspection,
with two anomalous data rows removed to mitigate
their distorting effects. This preprocessing resulted in
a refined dataset containing 978 validated
observations for each attribute.
3 INDICATORS SELECTION
The dependent variable of the multiple linear
regression model constructed by this research is the
price of smartphones. The predictor variables include
the average customer rating, the number of cores in
the processor, the processing speed of the processor,
etc. Table 1 shows the naming and interpretation of
the variables:
Table 1: The naming and explanation of variables
Variables Ex
p
lanation of Indicators
y
The price of the smartphone in USD
x
The average customer rating
x
The number of cores in the processor
x
The processing speed of the processo
r
x
The battery capacity
x
The wattage of the fast-charging feature
x
The amount of RAM
x
The internal storage capacit
y
x
The diagonal screen size
x
Indicates whether the phone supports
ex
p
andable stora
g
e
x
The height of the screen resolution
x
The width of the screen resolution
x
Indicates whether the phone has 5G support
or not
x
The screen refresh rate
x
The number of rear cameras
x
The resolution of the primary rear camera
x
The resolution of the front camera
IAMPA 2025 - The International Conference on Innovations in Applied Mathematics, Physics, and Astronomy
66
Since insignificant variables would reduce the
goodness of fit of the model, this study employed
backward stepwise regression to gradually eliminate
the insignificant variables, thereby enhancing the
reliability of the model. Set the significance level to
0.05, Table 2 lists the excluded variables:
Table 2: The Excluded Variable
x
x
x
x
x
p
0.839 0.707 0.526 0.838 0.217
As can be seen from Table 2, the P values of the
five
variables in the table are all greater than 0.05, which
indicates that the influence of x
, x
, x
, x
, x
is not significant and thus has been excluded.
4 MODEL CONSTRUCTION AND
REGRESSION ANALYSIS
Based on the backward stepwise regression method,
factors significantly influencing the price of smart
phones are selected to construct a multiple linear
regression model. The regression equation is as
follows:
y
76.027 x
78.904 x
220.091 x
0.036 x
1.885 x
9.899x
1.532 x
98.088
x
105.903x
0.100x
0.205x
ε
(1)
The regression analysis of the model was
conducted using R software. With the significance
level set at 0.05, the regression results are presented
in Table 3.
Table 3: Linear regression analysis results
B Standard erro
r
Beta t
p
x
-76.027 10.966 -1.154 -6.933 0.000
x
-78.904 9.216 -1.192 -8.562 0.000
x
220.091 22.823 1.048 9.643 0.000
x
-0.036 0.007 -0.344 -5.045 0.000
x
-1.885 0.271 -0.194 -6.954 0.000
x
9.899 3.868 0.136 2.559 0.011
x
1.532 0.080 0.524 19.0580.000
x
98.088 15.990 1.241 6.135 0.000
x
-105.903 19.637 -0.163 -5.393 0.000
x
0.100 0.017 0.441 5.850 0.000
x
0.205 0.026 0.442 7.965 0.000
R² 0.845
Ad
j
usted R² 0.843
F F
(
11,967
)
=478.891,
p
=0.000
According to the data analysis in Table 3, the
processing speed of the processor(x3), the amount of
RAM(x6), and the internal storage capacity(x7) have
a positive impact on smartphone prices, which
reflects the significance of processor performance
and storage capacity in determining the prices of
smartphones. Similarly, Ahmad, Ahmed, & Ahmad
(2019) also confirmed that storage capacity plays an
important role in determining the price of
smartphones.
Meanwhile, the screen size(x8) is also positively
correlated with the smartphone prices. This might be
because larger screen sizes lead to higher costs for
smartphones, and large-screen phones have
advantages in terms of visual experience and
operational space. Tanveer et al. (2021)'s research
results are in line with this finding. Moreover, the
screen resolution(x10, x11) also has a significant
positive impact on prices, which might indicate that
consumers prefer smartphones with clearer screens.
Obviously, this study also has some key factors
that are negatively correlated with the price of
smartphones. Specifically, the average customer
rating(x1) has a negative correlation with the
smartphone prices. This discovery reflects the trend
that the higher the average rating of a smartphone, the
lower its price. This might imply that consumers
prefer smartphones with better value for money. It is
also worth noting that the number of cores in the
processor(x2) and the battery capacity(x4) also have
a negative impact on the smartphone prices.
Generally speaking, the more cores a smartphone
processor has or the larger its battery capacity is, the
heavier the smartphone will be. This outcome implies
that consumers may prefer the slim and portable
models. In addition, the wattage of the fast-charging
feature(x5) is also in an inverse relationship with the
smartphone prices, which might reflect that there is a
trend for low-priced smartphones to adopt
fast-charging technology. Finally, whether the
smartphone supports expandable storage(x9) is also
negatively correlated with its price. Mid-range and
low-end smartphones often enhance their
cost-effectiveness and appeal by supporting storage
expansion. Chen (2023) reported a similar
conclusion.
5 ADAPTIVE TEST OF THE
MODEL
Table 4: Model summary
R R² Ad
j
usted R² RMSE MAE RMR
0.9190.845 0.843 203.812 132.403205.074
Analysis of Factors Influencing Smartphone Prices
67
As can be seen from Table 4, the correlation
coefficient R is 0.919, indicating a strong positive
correlation between these influencing factors and the
price of smartphones. The coefficient of
determination R² is 0.845, which indicates that this
model can account for 85.3% of the variation in
smartphone prices. Adjusted R² is an adjustment to R²,
and its value is 0.843, indicating that this model has a
relatively good goodness of fit. It is also worthy of
attention that the values of Root Mean Square Error
(RMSE), Mean Absolute Error (MAE), and Relative
Mean Residual (RMR) are relatively small,
indicating that the prediction accuracy of this model
is relatively high.
Figure 1 presents the standardized residual plot
and P-P plot. From Figure 1, it can be observed that
the distribution of residuals is basically per the
normal distribution, which indicates that the
reliability of the model is relatively high.
Figure 1: Residual diagnosis chart (Photo/Picture credit:
Original).
Table 5: ANOVA form
Sum of squares df
Mean
s
q
uare
F
p-valu
e
Regressi
on
221310274.906 11
20119115.
901
478.891 0.000
Residual
erro
r
40625502.027 967 42011.895
Sum of squares df
Mean
s
q
uare
F
p-valu
e
Total 261935776.933 978
As shown in Table 5, the model passed the F test
(F=478.891, p=0.000<0.05), which indicates that the
construction of the model is meaningful. Therefore,
the multiple linear regression model constructed in
this study has a very good explanatory power for
understanding how different characteristics affect the
prices of smartphones.
6 CONCLUSION
Through linear regression analysis, this study has
revealed the key factors influencing the prices of
smartphones. The positive impacts of the processor
speed, the amount of RAM, the internal storage
capacity, the screen size and the screen resolution
indicate that consumers prefer phones with superior
performance and are willing to pay higher prices for
them. On the other hand, the negative impacts of the
average customer rating、the number of cores in the
processor, the battery capacity, charging power and
the support for storage expansion by smartphones
reveal another important feature of the current
smartphone market. That is, when consumers
purchase smartphones, they pay more attention to the
overall practicality and cost-effectiveness of the
devices rather than the mere accumulation of single
functions. It is worth noting that although the model
constructed in this study has a relatively good
explanatory power, its R² value is less than 1. This
indicates that there are still other unaccounted
influencing factors that need to be captured. Future
researchers should adopt a more comprehensive
approach to construct a new model.
Based on the findings of this study, smartphone
manufacturers should focus their attention on the core
functions of their products and formulate reasonable
pricing strategies to meet consumers' pursuit of
cost-effectiveness and practicality of functions.
Meanwhile, manufacturers should avoid overloading
the product configuration. Instead, they should attract
consumers by optimizing product design and
functional combination, thereby maximizing profits.
AUTHOR CONTRIBUTIONS
All the authors of this article have made the same
contribution. The names of the authors are listed in
the order of the initial letters of their surnames.
IAMPA 2025 - The International Conference on Innovations in Applied Mathematics, Physics, and Astronomy
68
REFERENCES
Ahmed, T., Ahmad, W. & Bashir, A. 2022. Assessing the
value of 3G and 4G network modes in mobile phone
pricing. Digital Policy Regulation and Governance,
24(4), 380-397.
Ahmad, W., Ahmed, T. & Ahmad, B. 2019. Pricing of
mobile phone attributes at the retail level in a
developing country: Hedonic analysis.
Telecommunications Policy, 43(4), 299-309.
Chen, J. L. 2023. How will the price hikes by storage
manufacturers by 20% affect downstream mobile
phone manufacturers. China Business News, C02.
Chen, K. X. 2024. Improvement of the method for
compiling the characteristic price index and prediction
of the price index. Master's Thesis, Inner Mongolia
University of Finance and Economics.
Dahatonde, A. 2023. Real-world smartphones dataset.
https://www.kaggle.com/datasets/abhijitdahatonde/real
-world-smartphones-dataset
Han, H. G., Li, Y. & Du, Y. P. 2022. Adaptive price
adjustment method for used mobile phone based on
dual deep fuzzy networks. Science China
Technological Sciences, 65(06), 1330-1337.
Liu, J. & Mo, Z. 2021. The effects of review’s mobile
phone price on consumers’ purchase intention: An
event-related potential study. Journal of Neuroscience
Psychology and Economics, 14(4), 197-206.
Maulud, D. & Abdulazeez, A. 2020. A review on linear
regression comprehensive in machine learning. Journal
of Applied Science and Technology Trends, 1, 140-147.
Qin, X. H. & Ren, S. J. 2021. Pricing strategy for new
smartphones. Cooperative Economy and Technology,
09, 70-71.
Tanveer, M., Kaur, H., Thomas, G., Mahmood, H., Paruthi,
M. & Yu, Z. 2021. Mobile phone buying decisions
among young adults: An empirical study of influencing
factors. Sustainability, 13(19), 10705.
Wang, X. Y. & Wang, Y. 2022. Research on marketing
strategies for domestic smartphones. Modern
Marketing Special Issue, 02, 42-44.
Xu, L. X. 2022. Research on classification and prediction
of mobile phone prices based on machine learning.
Master's Thesis, Chongqing University.
Analysis of Factors Influencing Smartphone Prices
69
