Exploring the Influencing Factors of Automobile Prices Based on the

Dataset for Vehicle Price Prediction

Boyu Wang

School of Data Science and Artificial Intelligence, Dongbei University of Finance and Economics, Dalian,

Liaoning,116025, China

Keywords: Car Price, Production Year, Engine Displacement, Mileage.

Abstract: With the continuous development of the automotive market, the research on the influencing factors of car

prices plays a crucial role in industry decision-making and consumer guidance. However, previous studies on

the factors affecting car prices have only remained at the level of single variable analysis, lacking a systematic

analysis of the combined influence of multiple variables, making it difficult to provide a scientific basis for

market behavior. Therefore, this paper aims to study the influence mechanism of key variables such as

production year, engine displacement, and mileage on car prices. This study uses descriptive statistics,

correlation analysis, and multiple linear regression models to quantify the impact of each variable on car

prices. To address the issue of insufficient variable correlation analysis, correlation analysis is used to explore

the initial connections between variables. Further, through multiple linear regression models, the specific

impact of each factor on price is clarified. Based on 10,000 car data, the model was constructed, and it was

found that the production year has the most significant impact on price, followed by engine displacement and

mileage, with the model's explanatory power reaching 87.6%. The number of doors and the number of owners

have no statistically significant impact on price. These conclusions provide consumers with a "year + mileage"

dual-core evaluation framework.

1 INTRODUCTION

In the automotive market, the influencing factors of

car prices have always been the focus of attention for

consumers, automakers, and dealers. Understanding

these factors not only helps consumers make wiser

decisions when purchasing a car but also enables

automakers to formulate more competitive pricing

strategies and provides a basis for dealers to optimize

inventory management and cost control. Therefore,

exploring the influencing factors of car prices has

significant practical significance.

Previous studies have identified multiple key

factors that affect car prices. Firstly, brand is one of

the important factors influencing car prices. Luxury

brands such as BMW and Mercedes-Benz, with their

high quality and good market reputation, can usually

sell at higher prices (Wu, 2025). Secondly, different

models within the same brand also show significant

price differences. For instance, Toyota's Corolla, as a

family car for the general public, is relatively

https://orcid.org/0009-0004-4875-0967

affordable, while the Prado, as a mid-to-large SUV

with higher configurations, has a correspondingly

higher price (Luo, 2025). Additionally, the

production year also has a significant impact on car

prices. With technological advancements, newly

produced cars typically adopt more advanced

technologies and manufacturing processes, offering

better performance and safety, and thus command

higher prices (Zhao, 2023).

Engine size is another crucial factor. Cars

equipped with larger engines usually have stronger

power, but due to their higher manufacturing

difficulty and cost, they tend to be more expensive

(Deng et al., 2022). Meanwhile, with the increasing

awareness of environmental protection and the

innovation of energy technologies, the type of fuel

has become an increasingly significant factor

affecting car prices. Traditional gasoline and diesel

vehicles have obvious price differences due to fuel

costs and technological differences. However, hybrid

and electric vehicles, due to their high research and

600

Wang, B.

Exploring the Inﬂuencing Factors of Automobile Prices Based on the Dataset for Vehicle Price Prediction.

DOI: 10.5220/0013833900004708

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 600-604

ISBN: 978-989-758-774-0

development costs and rapid updates in battery

technology, experience greater price fluctuations

(Zhao, 2023). Moreover, the type of transmission also

affects car prices. Cars with automatic transmissions

are generally more expensive than those with manual

transmissions because automatic transmissions better

suit the driving habits of most consumers (Zhang,

2022).

Although previous studies have achieved certain

results in identifying the factors influencing car prices,

they generally suffer from the limitations of single

data sources and limited sample sizes, leading to an

insufficiently comprehensive coverage of scenarios

(Chen, 2023). Moreover, the existing analytical

methods are mostly simple correlation analyses,

failing to deeply explore the complex interactions

among various factors, leaving room for further

research.

To break through the limitations of previous

studies, this paper mainly uses correlation analysis

and multiple linear regression (MLR) analysis

methods to explore the influence of various factors on

car prices, reveal their intrinsic relationships, and

provide scientific decision-making bases for all

parties in the automotive market, promoting its

healthy development.

2 METHODS

2.1 Data Sources and Description

This paper is based on a car price prediction dataset

containing 10,000 entries from the Kaggle website,

which covers various aspects of information such as

brand, model, production year, engine size, fuel type,

transmission type, mileage, number of doors, and the

number of previous owners (Mustafa, 2025; Wang,

2023). Before the analysis, this study cleaned and

preprocessed the data to ensure its reliability. All the

data will be used in the subsequent research.

2.2 Selection and Explanation of

Indicators

Table 1 presents the selection and explanation table

of key indicators affecting car prices, listing seven

variable names such as year and engine size, and

briefly explaining the principles by which each

variable affects car prices.

Table 1: Key Indicators Selection and Explanation for Factors Affecting Automobile Prices

Number Variable Name Brief Description

Year New cars are usually more expensive

Engine Size Large engines are costly, making the car price higher

Mileage The higher the mileage, the lower the price usually is

Doors Different door numbers represent different models, and the prices vary

Owner Count Frequent changes of ownership lead to a lower price

2.3 Method Introduction

This study employs descriptive statistics, correlation

analysis, and MLR models for analysis. Descriptive

statistics are used to summarize the characteristics of

the data, revealing the central tendency and

dispersion of variables through the calculation of

means, standard deviations, and other indicators,

providing a foundational framework for subsequent

analysis. Correlation analysis, based on Pearson's

correlation coefficient, quantifies the strength and

direction of linear relationships between variables,

helping to identify those significantly related to car

prices and select core independent variables for

model construction. The MLR model is constructed

with car price as the dependent variable and x1, x2,

x3, x4, and x5 as independent variables, with the

formula as follows:

Price = β



+β



+β



+β



+β





+ε (1)

This method can control the influence of other

variables and independently assess the marginal

effect of each independent variable on the price. The

advantage of the model lies in its strong

interpretability, allowing for a direct comparison of

the degree of variable influence through standardized

coefficients. This method is particularly suitable for

the quantitative analysis of the combined influence of

multiple variables, breaking through the limitations of

single-variable analysis.

Exploring the Inﬂuencing Factors of Automobile Prices Based on the Dataset for Vehicle Price Prediction

601

3 RESULTS AND DISCUSSION

3.1 Descriptive Statistical Analysis

Through descriptive statistics of the variables, the key

results in Table 2 were obtained:

Table 2: Descriptive Statistics of Variables

Variable Case Number Average Value Standard Deviation

Effective Missing

Year 10000 0 2011.540 6.898

Engine Size 10000 0 3.001 1.149

Mileage 10000 0 149239.110 86322.349

Doors 10000 0 3.500 1.110

Owner Count 10000 0 2.990 1.423

Price 10000 0 8852.96 3112.597

Table 2 indicates that the production years of the

vehicles are mainly concentrated around 2011, with a

relatively concentrated time distribution. The average

engine displacement is 3.001, mainly medium

displacement. The standard deviation of mileage is as

high as 86,322.349, indicating significant differences

in vehicle mileage; the number of doors is mainly

concentrated between 3 and 4. The number of vehicle

owners fluctuates, but is mainly distributed around 3

people. The price of cars has a high degree of

dispersion and significant individual differences.

3.2 Correlation Analysis of Variables

and Prices

Through the correlation analysis of variables, the key

results in Table 3 were obtained:

Table 3: Correlation Coefficient Table of Key Variables Affecting Automobile Prices

Relevance

Year Engine Size Mileage Doors Owner Count Price

Year 1

Engine Size -0.001 1

Mileage -0.002 0.015 1

Doors 0.015 -0.01 0.008 1

Owner Count 0 0.007 0.006 -0.005 1

Price 0.663** 0.357** -0.551** 0.001 0.003 1

** At the 0.01 level (two-tailed), the correlation is significant.

Table 3 indicates that the production year is

significantly positively correlated with the car price;

the newer the year, the higher the price. Engine

displacement is significantly positively correlated

with the price; the larger the displacement, the higher

the price. Mileage is significantly negatively

correlated with the price; the more mileage, the lower

the price. However, the correlation coefficients of the

number of doors and the number of previous owners

with the car price are close to 0 and do not show a

significant correlation, indicating that these two

variables have little impact on the car price.

3.3 MLR Analysis

This study constructed an MLR model to explore the

key factors influencing car prices. The dependent

variable of the model is the car price, and the

independent variables include: production year,

engine size, mileage, number of doors, and number of

previous owners. The results are shown in Table 4:

IAMPA 2025 - The International Conference on Innovations in Applied Mathematics, Physics, and Astronomy

602

Table 4: Results of MLR Analysis of Factors Affecting Car Prices

Linear regression analysis results (n=10000)

Unstandardized coefficient

Standardized

coefficient

t p

β standard error Beta

Constant -592198.767 3196.549 - -185.262 0.000**

Year 298.801 1.589 0.662 188.033 0.000**

Engine Size 992.914 9.538 0.367 104.105 0.000**

Mileage -0.020 0.000 -0.555 -157.636 0.000**

Doors -3.732 9.875 -0.001 -0.378 0.705

Owner_Count 7.551 7.704 0.003 0.980 0.327

R² 0.876

Adjusted R² 0.876

F F (5,9994)=14132.594, p=0.000

Note: Dependent variable = Price

* p<0.05 ** p<0.01

As shown in Table 3, with price as the dependent

variable, year and engine size have a significant

positive impact on price (p < 0.01), with standardized

coefficients of 0.662 and 0.367 respectively; mileage

has a significant negative impact on price (p < 0.01),

with a standardized coefficient of -0.555; the number

of doors and the number of owners have no

significant impact on price (p values are 0.705 and

0.327 respectively), with standardized coefficients

close to 0. The model's R² is 0.876, and the adjusted

R² is also 0.876, indicating that the model explains

87.6% of the price variation, validating the model's

effectiveness and suggesting that the model accounts

for approximately 87.6% of the price variation, with

a very good fit. The F-statistic value is 14132.594,

with a p-value < 0.01, indicating that the model as a

whole is significant. In summary, year, engine size,

and mileage are the main factors affecting price. The

prediction formula is:

Price = −592198.767 + 298.801x



992.914x



− 0.020x



− 3.732x



+ 7.551x



(2)

3.4 Limitations Analysis

This study has certain limitations in analyzing the

factors influencing car prices. Although the

introduction emphasized the significant role of brand

and model in determining prices, in the actual

modeling process, due to the high diversity of brands

and models in the dataset leading to an excessive

number of categorical variables and insufficient

sample sizes for some niche brands or unpopular

models, to control the complexity of the model, these

variables were ultimately not included in the MLR

analysis(Hao, 2025). This means the model failed to

fully reflect the impact of brand premiums and the

differences between popular and niche models on

prices, which may lead to deviations in the

interpretation of real market prices.

Furthermore, the study only explored the

relationship between variables based on a linear

model, while in reality, car prices may be influenced

by nonlinear factors or interaction effects. For

instance, the impact of engine displacement on price

may vary depending on brand positioning, and the

depreciation effect of mileage may show nonlinear

attenuation over the years. Additionally, fuel type and

transmission type were not deeply explored as

categorical variables in the regression analysis, which,

to some extent, simplified the complex influence

mechanisms of their technical costs and market

acceptance on prices.

The singularity of the data source also limits the

coverage of the sample in terms of the range of car

models and regional market environments, resulting

in insufficient explanatory power for niche brands or

special models (Wu et al., 2025). Future research

could attempt to include brand and model as

categorical variables, adopt nonlinear models, or

combine more diverse data sources to more

comprehensively reveal the price influence

mechanism under the interaction of multiple variables.

4 CONCLUSION

This study, through descriptive statistics, correlation

analysis, and MLR models, found that production

year, engine displacement, and mileage are the main

factors affecting car prices. The newer the year, the

Exploring the Inﬂuencing Factors of Automobile Prices Based on the Dataset for Vehicle Price Prediction

603

larger the displacement, and the lower the mileage,

the higher the price, with the year having a relatively

more significant impact. The number of doors and the

number of previous owners have no significant

impact on price. The model's explanatory power

reached 87.6%, indicating that the selected variables

can effectively explain price variations. The research

results provide data support for consumers' car

purchase decisions, car manufacturers' pricing

strategies, and dealers' inventory management.

However, the study has certain limitations. Due to

the characteristics of the data and the complexity of

the model, variables such as brand and model that

may have significant impacts on prices were not

included in the analysis, and the differences in brand

premiums and model positioning were not fully

reflected. Only a linear model was used to examine

the relationship between variables, which may

overlook nonlinear effects or interaction effects. The

sample coverage of a single dataset is limited, and its

explanatory power for niche models or specific

markets is insufficient. Future research could include

brand and model variables, combine nonlinear

models or interaction terms, and more finely depict

the mechanism of multi-variable interaction, and

introduce more diverse data sources to enhance the

applicability of the conclusions to complex market

scenarios.

REFERENCES

Chen, J., 2023. Research on the Development of New

Energy Vehicles and Charging Demand Management

(Master’s Thesis). Beijing Jiaotong University, Beijing.

Deng, Y., Li, J., Ren, H., & Jiao, Q., 2022. Slow cook-off

characteristics of HTPB engines with different

structural sizes. In Journal of Energetic Materials,

30(2), 155-162.

Hao, X. J. 2025. Reflection on tax reform to promote

comprehensive green transformation. Fiscal Science,

(01), 91-101.

Luo, K., 2025. Analysis of Vehicle Recalls in 2024:

Exceeding Ten Million Again After Five Years. In

China Quality Herald, 1, 44.

Mustafa, Ö., 2025. car_price: A Car Price Dataset with 10,

000 Variables. In Kaggle. https://www.kaggle.com/dat

asets/mustafaoz158/car-price

Wang, J., 2023. Research on Two-Stage Data Pricing for

Data Market Transaction Volume (Master's Thesis).

Zhongnan University of Economics and Law, Wuhan.

Wu, B., 2025. How Can Automakers Attract Consumers

Beyond Price Wars?. In China Consumer News, (003).

Wu, Z., Yang, J., Zhang, F., & Chen, Z. 2024. Exploring

automobile brand power: a multi-dimensional approach.

Environment, Development and Sustainability, 1-34.

Zhang, L., 2022. Research on the Used-Car Price Problem

Based on Semi-parametric Models (Master's Thesis).

Chongqing Technology and Business University,

Chongqing.

Zhao, Y., 2023. Research on Combustion and Emission

Performance of Ignition-type M100 Methanol Engine

(Master’s Thesis). Taiyuan University of Technology,

Taiyuan.

IAMPA 2025 - The International Conference on Innovations in Applied Mathematics, Physics, and Astronomy

604