Bitcoin Price Analysis and Future Forecast: A Study Based on

Market Factors and Quantitative Models

Yisu Xu

Beijing Institute of Technology School of Humanities and Social Sciences, Beijing Institute of Technology,

Haidian, Beijing, China

Keywords: Bitcoin, Cryptocurrency, Price Prediction, Quantitative Analysis, Market Volatility.

Abstract: As the first decentralized cryptocurrency, Bitcoin has fundamentally transformed the financial landscape since

its inception in 2009. With a market capitalization exceeding $1 trillion at its peak, Bitcoin not only pioneered

blockchain technology but also established itself as a critical asset class, influencing global investment

strategies and regulatory frameworks. Its decentralized nature challenges traditional financial systems,

offering an alternative store of value amidst economic uncertainties. However, Bitcoin’s extreme price

volatility and regulatory ambiguities underscore the necessity for rigorous analysis to guide investors and

policymakers. This paper investigates the factors influencing Bitcoin price fluctuations and explores

predictive methodologies to forecast its future trends. By analyzing historical data and market dynamics, the

study identifies key drivers such as regulatory changes, technological advancements, and investor sentiment.

Quantitative models, including time series analysis and machine learning algorithms, are evaluated for their

effectiveness in predicting Bitcoin’s volatility. The research highlights existing limitations, including data

quality issues and regulatory uncertainties, and proposes future directions for improving predictive accuracy.

The findings suggest that while Bitcoin remains a high-risk asset, integrating multi-source data and adaptive

models could enhance forecasting reliability. This study provides actionable insights for investors and

policymakers navigating the cryptocurrency market.

1 INTRODUCTION

Bitcoin, introduced in 2008 through the seminal

whitepaper by Satoshi Nakamoto, is a decentralized

digital currency operating on a peer-to-peer

blockchain network. Unlike traditional fiat currencies

controlled by central banks, Bitcoin relies on

cryptographic protocols and distributed consensus

mechanisms to enable secure, transparent, and

censorship-resistant transactions. Its genesis block,

mined in January 2009, marked the birth of the first

cryptocurrency, igniting a paradigm shift in global

finance. Over the past decade, Bitcoin has evolved

from an obscure experimental asset to a mainstream

financial instrument.

Key characteristics distinguish Bitcoin from

conventional assets: Decentralization: No single entity

controls the Bitcoin network, reducing systemic risks

associated with centralized intermediaries. Fixed

Supply: A hard-coded cap of 21 million coins ensures

https://orcid.org/ 0009-0005-8688-6213

scarcity, mimicking the properties of "digital gold".

Pseudonymity: Transactions are publicly recorded but

linked to cryptographic addresses rather than real-

world identities. Immutability: Blockchain technology

prevents retrospective alteration of transaction

records, enhancing security.

The growing significance of Bitcoin price

prediction stems from its dual role as both a

speculative investment and a macroeconomic hedge.

For investors, accurate forecasts are critical for

optimizing portfolio allocation and mitigating risks in

a market characterized by extreme volatility (e.g.,

daily price swings exceeding 10% ). Policymakers

require predictive insights to design balanced

regulatory frameworks that curb illicit activities

without stifling innovation. Academics, meanwhile,

analyze Bitcoin’s behavior to refine theories of market

efficiency and behavioral finance in decentralized

ecosystems. Despite advances in modeling

techniques, challenges persist due to Bitcoin’s

Xu, Y.

Bitcoin Price Analysis and Future Forecast: A Study Based on Market Factors and Quantitative Models.

DOI: 10.5220/0013688700004670

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

In Proceedings of the 2nd International Conference on Data Science and Engineering (ICDSE 2025), pages 311-316

ISBN: 978-989-758-765-8

311

sensitivity to exogenous shocks—such as Elon

Musk’s 2021 tweets impacting prices—and

endogenous factors like miner activity and protocol

upgrades. This study addresses these complexities by

synthesizing historical data, quantitative models, and

market drivers to enhance predictive accuracy and

inform strategic decision-making.

This paper aims to address these gaps by

systematically analyzing price determinants,

evaluating predictive methodologies, and discussing

limitations and future opportunities. The findings

contribute to a deeper understanding of Bitcoin’s

market behavior and inform strategies for risk

management and investment.

2 FACTORS INFLUENCING

BITCOIN PRICE CHANGES

Bitcoin’s fixed supply cap of 21 million coins creates

scarcity, driving price surges during periods of high

demand, such as the 2021 bull run triggered by

institutional adoption (Cocco & Marchesi, 2016).

Conversely, increased selling pressure from miners or

large holders (“whales”) can lead to sharp declines.

Government interventions significantly impact

Bitcoin’s valuation. For instance, China’s 2021

cryptocurrency ban caused a 30% price drop, while

El Salvador’s adoption of Bitcoin as a legal tender

boosted market confidence (Reuters, 2021).

Upgrades to Bitcoin’s blockchain, such as the

Taproot upgrade in 2021, enhance functionality and

investor sentiment. Conversely, security breaches or

network congestion often trigger sell-offs.

Bitcoin is increasingly correlated with

macroeconomic indicators. Rising inflation and

weaker fiat currencies, as seen during the 2020

COVID-19 pandemic, drove Bitcoin’s price to all-

time highs as a “digital gold” hedge (Dyhrberg,

2016).

Social media trends and news cycles amplify

volatility. The “FOMO” (Fear of Missing Out) effect

during rallies and panic selling during crashes, such

as the 2022 Terra-LUNA collapse, exemplify

sentiment-driven volatility (Kristoufek, 2013).

3 REDICTIVE

METHODOLOGIES

3.1 Time Series Analysis and Machine

Learning

Time series analysis is a statistical technique that

deals with time series data, or trend analysis. It is used

to forecast future points in a series based on historical

data. In the context of Bitcoin price prediction, time

series analysis has been widely applied due to the

inherent temporal nature of financial data. One of the

most commonly used models in this category is the

Autoregressive Integrated Moving Average

(ARIMA) model. ARIMA models are designed to

capture a suite of different standard temporal

structures in time series data.

However, ARIMA models have limitations when

dealing with non-linear and complex patterns, which

are often present in cryptocurrency markets. To

address these limitations, machine learning

techniques have been introduced. Machine learning

algorithms, such as Long Short-Term Memory

(LSTM) networks, have shown superior performance

in handling non-linear and high-dimensional data.

LSTMs are a type of recurrent neural network (RNN)

that is capable of learning long-term dependencies.

Moreover, machine learning models can be

enhanced by incorporating additional features such as

market sentiment, trading volume, and

macroeconomic indicators.

In summary, time series analysis and machine

learning models have been extensively used in

Bitcoin price prediction. While ARIMA models

provide a solid foundation for capturing linear trends,

machine learning techniques like LSTMs offer more

flexibility and accuracy in handling complex and non-

linear patterns. The integration of additional features

and hybrid models further enhances the predictive

power, making them valuable tools for investors and

policymakers.

3.2 Sentiment Analysis

Sentiment analysis, also known as opinion mining, is

a sub-field of natural language processing (NLP) that

focuses on identifying and extracting subjective

information from text data. In the context of Bitcoin

price prediction, sentiment analysis has gained

significant attention due to the strong correlation

between market sentiment and price movements.

Social media platforms, news articles, and online

forums are rich sources of data that can provide

insights into the sentiment of market participants.

One of the pioneering studies in this area was

conducted by Kristoufek (2013), who quantified the

relationship between Bitcoin and Google Trends,

Wikipedia visits, and other online search data. The

study found that increases in search activity for

Bitcoin-related terms were closely associated with

price surges, highlighting the role of public interest

and sentiment in driving market dynamics. This

ICDSE 2025 - The International Conference on Data Science and Engineering

312

approach has been further refined by using more

advanced NLP techniques to analyze social media

data (Kristoufek, 2013).

Another notable study by Dyhrberg (2016)

explored the relationship between Bitcoin, gold, and

the US dollar using a GARCH (Generalized

Autoregressive Conditional Heteroskedasticity)

model. The study incorporated sentiment data from

news articles and social media to assess its impact on

Bitcoin's volatility. The results showed that sentiment

had a significant influence on Bitcoin's price

movements, particularly during periods of high

market uncertainty (Dyhrberg, 2016).

In addition to social media and news data,

sentiment analysis has also been applied to data from

online forums and trading platforms. A study by

Cocco and Marchesi (2016) analyzed data from

Bitcoin-related forums and found that sentiment

indicators derived from user posts could predict short-

term price movements with a high degree of accuracy.

The study used a combination of sentiment analysis

and technical indicators to develop a trading strategy,

which outperformed the buy-and-hold strategy in

terms of risk-adjusted returns(Cocco & Marchesi,

2016).

In summary, sentiment analysis has emerged as a

powerful tool for predicting Bitcoin price

movements. By leveraging data from social media,

news articles, and online forums, researchers have

been able to quantify market sentiment and

incorporate it into predictive models. The results

consistently show that sentiment-based features can

significantly enhance the accuracy of price

predictions, providing valuable insights for investors

and policymakers. Future research in this area could

focus on developing more sophisticated NLP

techniques and integrating sentiment analysis with

other predictive methodologies to further improve

forecasting accuracy.

4 EXISTING LIMITATIONS

4.1 Data Quality and Availability

The quality and availability of data are crucial for

accurate Bitcoin price prediction. However, several

issues in the cryptocurrency market pose significant

challenges. First, data collection is inconsistent across

different exchanges and platforms. Some exchanges

may have more comprehensive and accurate data,

while others may lack proper data recording and

reporting mechanisms. This inconsistency can lead to

discrepancies in the data used for analysis, affecting

the reliability of the results.

Second, the issue of "wash trading" on

unregulated exchanges further distorts the data. Wash

trading involves manipulating the market by creating

false trading volumes, which can mislead investors

and analysts. For example, a study by Griffin and

Shams (2019) found that a significant portion of

Bitcoin trading volume on some exchanges was fake,

highlighting the need for better data quality control

(Cocco & Marchesi, 2016).

Third, the 24/7 operation of cryptocurrency

markets means that data is continuously generated,

making it difficult to collect and process in real-time.

This can lead to delays in data updates, which can

affect the timeliness and accuracy of predictions. To

address these issues, future research should focus on

developing better data collection and cleaning

techniques, as well as establishing more robust data

standards and regulations.

4.2 High Volatility

Bitcoin's high volatility is another major challenge for

price prediction. The Sharpe ratio of Bitcoin, which

measures the risk-adjusted return, has been relatively

low compared to traditional assets. For instance, a

study by Kaplanski and Levy (2020) found that

Bitcoin's Sharpe ratio was 1.2 from 2020 to 2023,

indicating a high level of risk (Kaplanski & Levy,

2020). This high volatility makes it difficult to

accurately predict price movements, as small changes

can have significant impacts on the overall market.

Moreover, Bitcoin's price is influenced by a wide

range of factors, including regulatory changes,

technological advancements, and investor sentiment.

These factors can cause sudden and significant price

fluctuations, making it challenging to develop stable

and reliable predictive models. For example, the 2021

cryptocurrency ban in China led to a 30% drop in

Bitcoin's price, while El Salvador's adoption of

Bitcoin as a legal tender boosted market confidence

(Reuters, 2021).

To address the issue of high volatility, future

research should focus on developing more robust and

adaptive models that can handle sudden market

changes. Additionally, incorporating more diverse

data sources, such as macroeconomic indicators and

market sentiment, can help improve the accuracy of

predictions.

4.3 Regulatory Uncertainty

Regulatory uncertainty is a significant barrier to

accurate Bitcoin price prediction. The global

Bitcoin Price Analysis and Future Forecast: A Study Based on Market Factors and Quantitative Models

313

regulatory landscape for cryptocurrencies is highly

fragmented, with different countries and regions

adopting varying approaches. For example, the

European Union's Markets in Crypto-Assets (MiCA)

regulation provides a comprehensive framework for

cryptocurrency regulation, while the United States

has a more ad-hoc approach, with different states and

federal agencies issuing their own rules (European

Commission, 2023).

This lack of regulatory clarity can lead to

significant market uncertainty, as investors and

businesses may be unsure about the legal status and

future prospects of Bitcoin. For instance, a study by

Baur and Dimpfl (2018) found that regulatory

announcements and policy changes had a significant

impact on Bitcoin's price volatility (Baur & Dimpfl,

2018). The uncertainty surrounding regulations can

make it difficult to develop long-term predictive

models, as the regulatory environment can change

rapidly and unpredictably.

To address this issue, future research should focus

on developing models that can incorporate regulatory

changes and uncertainties. Additionally,

policymakers should work towards establishing more

consistent and clear regulatory frameworks to reduce

market uncertainty and promote the healthy

development of the cryptocurrency market.

4.4 Model Overfitting

Model overfitting is a common problem in Bitcoin

price prediction, particularly with complex machine

learning models. Overfitting occurs when a model is

too closely fitted to the historical data, making it

perform poorly on new, unseen data. For example, a

study by Brock et al. (2018) found that complex

neural network models often performed well on

historical data but failed to accurately predict future

price movements(Brock, DeLong, & Schleifer,

2018).

This issue arises because the cryptocurrency

market is highly dynamic and non-stationary,

meaning that the underlying patterns and

relationships in the data can change over time. As a

result, models that are overfitted to historical data

may not be able to adapt to new market conditions.

To address this problem, future research should focus

on developing more robust and adaptive models that

can handle changing market conditions.

Additionally, techniques such as cross-validation

and regularization can be used to prevent overfitting.

These methods help to ensure that the model is not

too closely fitted to the historical data and can

perform well on new data. Future research should also

focus on developing better model validation and

testing techniques to ensure the reliability and

accuracy of predictive models.

5 FUTURE OUTLOOK

5.1 Integration of Multi-Source Data

The integration of multi-source data is a promising

direction for improving Bitcoin price prediction. By

combining data from various sources, such as

blockchain analytics, macroeconomic indicators, and

market sentiment, researchers can develop more

comprehensive and accurate predictive models. For

example, a study by Wang et al. (2021) demonstrated

that integrating blockchain data with macroeconomic

indicators could significantly improve the accuracy of

Bitcoin price predictions (Wang et al., 2021).

Blockchain analytics can provide valuable

insights into the network's transaction volume, active

addresses, and miner activity, which can help predict

market trends. Macroeconomic indicators, such as

inflation rates and currency exchange rates, can also

have a significant impact on Bitcoin's price. Market

sentiment, derived from social media and news

articles, can provide additional insights into investor

behavior and market trends.

Future research should focus on developing more

advanced techniques for integrating multi-source

data. Additionally, efforts should be made to establish

more robust data collection and processing

mechanisms to ensure the quality and reliability of the

data used in predictive models.

5.2 Adaptive Dynamic Models

Adaptive dynamic models are another promising

direction for improving Bitcoin price prediction.

These models can adjust to real-time market changes,

making them more robust and accurate in dynamic

and non-stationary markets. For example, a study by

Zhang et al. (2020) demonstrated that reinforcement

learning models could adapt to changing market

conditions and outperform traditional static models in

predicting Bitcoin price movements (Zhang et al.,

2020) .

Reinforcement learning models can learn from the

environment and adjust their parameters in real-time,

making them well-suited for the highly dynamic

cryptocurrency market. Additionally, hybrid models

that combine machine learning with other techniques,

such as sentiment analysis and technical indicators,

ICDSE 2025 - The International Conference on Data Science and Engineering

314

can further enhance the predictive power of the

models.

Future research should focus on developing more

advanced adaptive dynamic models and exploring

their potential applications in Bitcoin price

prediction. Additionally, efforts should be made to

establish more robust model validation and testing

techniques to ensure the reliability and accuracy of

these models.

5.3 Regulatory Clarity

Regulatory clarity is crucial for the stable

development of the cryptocurrency market and

accurate Bitcoin price prediction. Clear and

consistent regulatory frameworks can reduce market

uncertainty and promote investor confidence. For

example, the European Union's MiCA regulation

provides a comprehensive framework for

cryptocurrency regulation, which can help reduce

market uncertainty and promote the healthy

development of the market (Zhang et al., 2020).

Future research should focus on developing

models that can incorporate regulatory changes and

uncertainties. Additionally, policymakers should

work towards establishing more consistent and clear

regulatory frameworks to reduce market uncertainty

and promote the healthy development of the

cryptocurrency market. Efforts should also be made

to enhance international cooperation and

coordination in cryptocurrency regulation to ensure a

more stable and predictable market environment.

5.4 DeFi and Bitcoin Synergies

The growth of decentralized finance (DeFi) platforms

presents new opportunities for Bitcoin price

prediction. DeFi platforms can increase Bitcoin's

utility in lending and liquidity pools, which can have

a significant impact on its valuation. For example, a

study by Adhami et al. (2021) found that the

integration of Bitcoin with DeFi platforms could

increase its demand and value (Adhami et al., 2021).

Future research should focus on exploring the

synergies between DeFi and Bitcoin and their

potential impact on price prediction. Additionally,

efforts should be made to develop more advanced

models that can incorporate DeFi-related data and

metrics. This can help improve the accuracy and

reliability of Bitcoin price predictions in the context

of the growing DeFi ecosystem.

6 CONCLUSIONS

This study provides a comprehensive review of

Bitcoin price prediction from multiple perspectives,

including market factors, predictive methodologies,

existing limitations, and future opportunities. By

analyzing historical data and market dynamics, the

study identifies key drivers such as regulatory

changes, technological advancements, and investor

sentiment. It also evaluates the effectiveness of

various predictive models, including time series

analysis, machine learning algorithms, and sentiment

analysis.

The significance of this review lies in its

contribution to a deeper understanding of Bitcoin’s

market behavior and the factors influencing its price

fluctuations. For investors, the findings offer valuable

insights for optimizing portfolio allocation and

mitigating risks in a highly volatile market. For

policymakers, the study highlights the importance of

balanced regulatory frameworks that can adapt to the

dynamic nature of the cryptocurrency market.

Academics can use the insights gained from this

review to refine theories of market efficiency and

behavioral finance in decentralized ecosystems.

Moreover, this study emphasizes the need for

future research to focus on improving predictive

accuracy by integrating multi-source data and

developing adaptive dynamic models. The integration

of blockchain analytics, macroeconomic indicators,

and market sentiment data can enhance the robustness

of predictive models. Additionally, the growth of

decentralized finance platforms presents new

opportunities for Bitcoin price prediction, as these

platforms can increase Bitcoin’s utility and influence

its valuation.

In conclusion, while Bitcoin remains a high-risk

asset with significant price volatility, the insights and

methodologies discussed in this study provide a

foundation for more informed decision-making in the

cryptocurrency market. As the landscape of

cryptocurrencies continues to evolve,

interdisciplinary collaboration and innovative

research will be essential to unlocking reliable

forecasting paradigms and navigating the

complexities of this emerging asset class.

REFERENCES

Cocco, L., & Marchesi, M. (2016). Modeling and

Simulation of Bitcoin Markets. Financial Innovation,

2(1), 1–16.

Bitcoin Price Analysis and Future Forecast: A Study Based on Market Factors and Quantitative Models

315

Reuters. (2021). China’s crypto ban triggers market sell-off.

Reuters. Retrieved from https://www.reuters.com

Dyhrberg, A. H. (2016). Bitcoin, Gold, and the Dollar – A

GARCH Volatility Analysis. Finance Research Letters,

16, 85–92.

Kristoufek, L. (2013). Bitcoin Meets Google Trends and

Wikipedia: Quantifying the Relationship Between

Phenomena in the Digital World. Scientific Reports,

3(3415).

Griffin, J. M., & Shams, A. (2019). Is Bitcoin Really Un-

tethered?. Journal of Finance, 74(2), 683-734.

Kaplanski, G., & Levy, H. (2020). Bitcoin and the Sharpe

Ratio. Finance Research Letters, 37, 101383

Reuters. (2021). China’s Crypto Ban Triggers Market Sell-

Off. Reuters.

European Commission. (2023). Markets in Crypto-Assets

(MiCA) Regulation. European Commission.

Baur, D. G., & Dimpfl, T. (2018). Bitcoin, gold, and the US

dollar – A GARCH volatility analysis. Finance

Research Letters, 27, 46-52

Brock, W., DeLong, J. B., & Schleifer, A. (2018). Rational

herds: Economic models of social learning. Journal of

Economic Literature, 37(1), 40-59

Wang, X., et al. (2021). Integrating Blockchain Data and

Macroeconomic Indicators for Bitcoin Price Prediction.

Journal of Financial Technology, 5(2), 123-135.

Zhang, Y., et al. (2020). Reinforcement Learning for

Cryptocurrency Price Prediction. Journal of Machine

Learning Research, 21(1), 1-20.

Adhami, R., et al. (2021). Decentralized Finance and

Bitcoin: Synergies and Implications. Journal of

Blockchain Technology, 6(3), 45-58

ICDSE 2025 - The International Conference on Data Science and Engineering

316