Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

17355269 · Jun 202619922001200920172026
48 results for cryptocurrency forecasting

Paper introduces probabilistic forecasting methods for cryptocurrency volatility.

problem Inadequate point forecasting methods for capturing full spectrum of volatility outcomes.
method Combines multiple base models (statistical and machine learning) to estimate conditional quantiles of cryptocurrency realized variance.
result QRS method outperforms sophisticated alternatives for Bitcoin volatility forecasting.

DAM improves cryptocurrency trend forecasting using multimodal data.

problem Simplistic merging of sentiment data in cryptocurrency trend forecasting.
method Dual Attention Mechanism (DAM) integrating financial metrics and sentiment analysis.
result DAM outperforms conventional models by up to 20% in prediction accuracy.

Cryptocurrency time-series predictability is low, resembling Brownian noise.

problem Low predictability of cryptocurrency exchange rates.
method Complexity and model predictions of Litecoin, Binance Coin, Bitcoin, Ethereum, and XRP exchange rates.
result Simpler models outperform complex ones in cryptocurrency forecasting.

Study shows diverse data sources improve cryptocurrency forecasting models.

problem Improving cryptocurrency market forecasting accuracy.
method Integrating various data types, including on-chain metrics, traditional indices, and macroeconomic indicators.
result Data source diversity significantly enhances forecasting model performance.

Cryptocurrency forecasting model considers macro, sentiment, and technical indicators.

problem High price volatility in cryptocurrency markets.
method Dual-prediction mechanism incorporating macroeconomic fluctuations, technical indicators, and individual cryptocurrency price changes.
result The proposed model outperforms ten comparison methods in short-term cryptocurrency forecasting.

FinBERT-BiLSTM predicts cryptocurrency prices using sentiment analysis.

problem Predicting volatile cryptocurrency market prices.
method Hybrid model combining Bi-LSTM and FinBERT for sentiment analysis.
result Enhanced forecasting accuracy for volatile financial markets.

The paper improves cryptocurrency price forecasting using deep learning and NLP on financial, blockchain, and social media data.

problem Improving cryptocurrency price forecasting accuracy and profitability.
method Integrates financial, blockchain, and social media data; applies BART MNLI model for sentiment analysis; uses deep learning NLP models; compares with traditional methods; uses local extrema as predictive targets.
result Significantly improves forecasting accuracy and profitability of cryptocurrency price predictions.

CryptoGAT improves cryptocurrency price prediction by treating it as a graph problem.

problem Cryptocurrency price prediction challenges due to extreme volatility.
method CryptoGAT, a Graph Attention Network, redefines cryptocurrency prediction as a cross-asset graph problem.
result CryptoGAT outperforms state-of-the-art methods in cryptocurrency price prediction.

Study evaluates deep learning models for cryptocurrency price prediction.

problem Accurate cryptocurrency price forecasting models are needed due to market volatility.
method Reviewed and evaluated deep learning models including LSTM, CNN, and Transformer.
result Convolutional LSTM with multivariate approach provides best prediction accuracy.

Temporal mixture ensemble predicts cryptocurrency exchange volumes better than traditional methods.

problem Intraday volume forecasting in cryptocurrency markets.
method Temporal mixture ensemble model using transaction and order book data.
result The model outperforms traditional time series and machine learning methods.

Combines VaR and ES forecasts for cryptocurrency market risk management.

problem Improving tail risk forecasts in financial markets.
method Proposes semiparametric and parametric combination frameworks.
result Combined forecasts outperform individual VaR and ES forecasts.

The study uses Markov chains to forecast cryptocurrency market dynamics.

problem Forecasting and understanding market fluctuations in cryptocurrencies.
method Markov chains of orders one to eight were used to forecast intra-day returns of three major cryptocurrencies.
result Predictions from empirical probabilities outperform random choices.

This paper forecasts cryptocurrency log-returns using LASSO-VAR and sentiment analysis.

problem Forecasting log-returns of cryptocurrencies using social media sentiment.
method LASSO-VAR model combined with Twitter and Reddit sentiment data.
result The model predicts the correct direction of cryptocurrency returns more than 50% of the time.

Study improves cryptocurrency price prediction using unlabeled text data.

problem Predicting cryptocurrency returns from unlabelled text data.
method Introduced weak learning approach to finetune BERT on unlabeled text data.
result Finetuning pretrained NLP models with weak labels enhances forecast accuracy.

This study improves cryptocurrency price forecasting using time series categorization and deep learning.

problem Accurate prediction of cryptocurrency prices is challenging due to limited data and diverse behaviors.
method The approach involves categorizing financial time series, creating deep learning models for each category, and combining data from other cryptocurrencies to increase training data.
result The method increases prediction accuracy by learning each subseries category with similar behavior and combining data from other cryptocurrencies.

This paper proposes new GARCH models for cryptocurrency volatility, showing skewed distributions improve prediction accuracy.

problem Predicting cryptocurrency volatility and improving upon normality assumptions.
method Non-Gaussian GARCH models with Skewed Generalized Error Distribution.
result Skewed distributions enhance forecasting accuracy for cryptocurrency exchange rates.

The paper assesses dimensionality reduction for cryptocurrency link prediction.

problem Establishing a link between cryptocurrencies using dimensionality reduction techniques.
method Used canonical correlation analysis and principal component analysis on log returns and covariates of Bitcoin and Ethereum.
result Performance of dimensionality reduction techniques in forecasting Ethereum returns with Bitcoin features.

Kalshi prediction markets forecast cryptocurrency volatility through monetary policy and inflation signals.

problem Forecasting cryptocurrency volatility using prediction markets.
method Monetary policy and inflation signals from Kalshi prediction markets.
result Signals from Kalshi prediction markets predict cryptocurrency volatility with statistical significance.

Transformer predicts Ethereum prices using cross-currency correlation and sentiment analysis.

problem Predicting Ethereum cryptocurrency prices with limited data.
method Transformer-based neural network with cross-currency correlation and sentiment analysis.
result Transformer model outperforms other models on some parameters.

This paper optimizes cryptocurrency portfolios by clustering price correlations and improving risk-return profiles.

problem Volatility and regulatory uncertainty in cryptocurrency markets make portfolio construction challenging.
method The paper combines network analysis, price forecasting, and portfolio theory to identify stable groups of correlated cryptocurrencies.
result Predictive consensus-clustering portfolios maintain positive and stable performance up to a 14-day horizon, with favourable gain-loss asymmetry and tighter tail-risk control.

Adaptive TFTs improve cryptocurrency price prediction accuracy.

problem Precise short-term price prediction in volatile cryptocurrency markets.
method Dynamic subseries lengths and pattern-based categorization.
result Significantly outperforms baseline models in prediction accuracy and profitability.

CTBench benchmarks cryptocurrency time series generation for trading applications.

problem Lack of comprehensive benchmarks for cryptocurrency time series generation.
method Developed a comprehensive benchmark extsf{CTBench} with 13 metrics across 5 dimensions.
result Uncovered trade-offs between statistical fidelity and real-world profitability.

Transformer models outperform LSTM in financial forecasting with MADL loss.

problem Optimizing loss functions for Transformer models in financial forecasting.
method Empirical experiments with MADL loss function on equity and cryptocurrency assets.
result Transformer models significantly outperform LSTM models in financial forecasting.

Paper introduces MADL loss function for better AIS model optimization.

problem Optimizing machine learning models for AIS construction.
method Proposes Mean Absolute Directional Loss (MADL) function.
result MADL function improves hyperparameter selection and investment strategy efficiency.

Study compares forecasting models for European financial markets and cryptocurrencies, finding hybrid ETS-ANN model best.

problem Challenges in predicting financial market fluctuations and cryptocurrency prices.
method Comparative analysis of ARIMA, hybrid ETS-ANN, and kNN models on European financial markets and cryptocurrency data.
result Hybrid ETS-ANN model performs best over extended periods, with moderate accuracy.

Study uses RNN for real-time crypto price prediction and trading optimization.

problem High volatility in cryptocurrency markets makes traditional forecasting models unreliable.
method Data collection, preprocessing, model refinement, and backtesting.
result Improved accuracy in real-time crypto price prediction and optimized trading strategies.

Study uses sentiment analysis to predict cryptocurrency token returns in virtual reality.

problem Predicting cryptocurrency token returns in virtual reality economies.
method Used BERT for sentiment analysis and developed LSTM models integrating multi-modal features.
result Multi-modal model significantly outperforms price-only baseline in prediction accuracy.

The paper predicts cryptocurrency prices using a path-dependent Monte Carlo simulation.

problem Forecasting cryptocurrency prices with volatility and jumps.
method Merton's jump diffusion model with machine learning, traditional, and statistical methods.
result Introduced a path-dependent Monte Carlo simulation for cryptocurrency price prediction.

The study improves Bitcoin price prediction using hybrid machine learning and enhances interpretability.

problem Improving Bitcoin price prediction accuracy and interpretability.
method Hybrid machine learning algorithms (OLS, LASSO, LSTM, decision tree regressors) and preprocessing techniques for time-series data.
result Linear regression achieves the best performance in predicting Bitcoin prices.

Social media signals have been successfully used to develop large-scale predictive and anticipatory analytics. For example, forecasting stock market prices and influenza outbreaks. Recently, social data has been explored to forecast price fluctuations of cryptocurrencies, which are a novel disruptive technology with si…

2019-07-01abs ↗pdf ↗

New framework predicts crypto volatility, outperforming traditional models.

problem Forecasting volatility in cryptocurrencies during the crypto-winter.
method Combines LSTM and rough volatility models, using a parsimonious parametric model.
result Similar prediction performances with fewer parameters, suggesting universality of volatility mechanisms.

Novel S-MF-DFA detects structured multifractality in crypto markets.

problem Analyzing scaling regularity of cryptocurrencies.
method Structural detrended multifractal fluctuation analysis (S-MF-DFA) with change-points detection.
result Main cryptocurrencies exhibit structured multifractality, with decreasing multifractality after 2018.

The study uses Bayesian Hidden Markov Models to predict cryptocurrency returns.

problem Predicting the volatility and trends of cryptocurrencies.
method Bayesian Hidden Markov Models with four states to capture different return characteristics.
result The NHHM model with four states outperforms other models in predicting cryptocurrency returns.