The paper proposes a new method to predict VaR using DCS and generalized distributions.
problem Improving VaR prediction models in financial risk management.
method Dynamic Conditional Score (DCS) model combined with generalized distributions (GD).
result The proposed model outperforms traditional models in high-risk VaR prediction.
The study challenges the reliability of VaR due to market randomness.
problem Reliability and accuracy of VaR predictions are compromised by market randomness.
method Introduces market-based probabilities of price and return, dependent on trade values and volumes.
result Market-based price volatility is more accurate than frequency-based VaR predictions.
Several well-established benchmark predictors exist for Value-at-Risk (VaR), a major instrument for financial risk management. Hybrid methods combining AR-GARCH filtering with skewed-t residuals and the extreme value theory-based approach are particularly recommended. This study introduces yet another VaR predictor, …
The study improves VaR forecast accuracy by modeling conditional quantile dynamics.
problem Improving the accuracy of Value-at-Risk (VaR) forecasts for time-varying quantiles.
method Time-varying modeling of VaR, evaluation via simulation, asymmetric Mean Absolute Deviation loss function.
result Substantial improvements in forecasting conditional quantiles by maintaining predicted quantile unchanged.
GAS models have been recently proposed in time-series econometrics as valuable tools for signal extraction and prediction. This paper details how financial risk managers can use GAS models for Value-at-Risk (VaR) prediction using the novel GAS package for R. Details and code snippets for prediction, comparison and back…
Causal-NECO VaR improves financial risk assessment under market turbulence.
problem Inaccurate risk assessment in volatile markets.
method Causal Network Contagion Value at Risk (Causal-NECO VaR) using causal network framework.
result Robust and invariant predictive power in unstable financial environments.
Bayesian approach improves portfolio optimization using VaR and CVaR.
problem Optimizing portfolio weights using VaR and CVaR for risk management.
method Bayesian perspective, posterior predictive distribution, observed data.
result Bayesian approach yields more accurate optimal portfolio weights.
Hybrid model combines VAR and neural network for OFI prediction.
problem Accurate prediction of Order Flow Imbalance (OFI) in high frequency trading.
method Combines Vector Auto Regression (VAR) and a simple feedforward neural network (FNN).
result Hybrid model achieves superior predictive accuracy compared to standalone models.
A new risk measure, the lambda value at risk (Lambda VaR), has been recently proposed from a theoretical point of view as a generalization of the value at risk (VaR). The Lambda VaR appears attractive for its potential ability to solve several problems of the VaR. In this paper we propose three nonparametric backtestin…
New method uses G-expectation for financial risk measurement.
problem Measuring uncertainty in financial time series.
method Introducing G-normal distribution, applying max-mean estimators, and using autoregressive models.
result G-VaR model outperforms other VaR predictors in risk prediction.
GRF models predict cryptocurrency VaR better than other methods.
problem Predicting Value at Risk (VaR) for volatile cryptocurrencies.
method Generalized Random Forests (GRF) adapted for quantile prediction.
result GRF models outperform other methods in cryptocurrency VaR predictions.
Bayesian approach confirms no return predictability for 1926-2004 data, weak evidence for 1953-2021.
problem Investigating return predictability using Bayesian methods.
method Developed a new shrinkage type prior for a model parameter in a VAR system, compared to other estimation methods.
result Bayesian approach outperforms reduced-bias estimator in terms of size and power.
RNN-HAR model improves VaR forecasting with long-memory and non-linear dynamics.
problem Efficiently forecasting Value at Risk (VaR) with long-memory and non-linear realized volatility.
method Loss-based generalized Bayesian inference with Sequential Monte Carlo for model estimation and prediction.
result RNN-HAR model consistently outperforms other VaR forecasting models.
Optimizes forecast distributions for financial risk management.
problem Improving risk management through better forecast distributions.
method Optimizes forecast distributions using scoring rules relevant to financial risk management.
result Tail-focused predictive distributions yield better outcomes in hedging strategies involving VIX futures.
Study finds no consistent return predictability using payout ratios across 16 countries.
problem Return predictability using payout ratios in various markets.
method Analysis of 16 developed countries' bond, equity, and housing markets using payout-price ratios.
result No consistent in-sample and out-of-sample performance with positive utility gain.
Multivariate time-series modeling and forecasting is an important problem with numerous applications. Traditional approaches such as VAR (vector auto-regressive) models and more recent approaches such as RNNs (recurrent neural networks) are indispensable tools in modeling time-series data. In many multivariate time ser…
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.
In the paper, we consider the problem of link prediction in time-evolving graphs. We assume that certain graph features, such as the node degree, follow a vector autoregressive (VAR) model and we propose to use this information to improve the accuracy of prediction. Our strategy involves a joint optimization procedure …
We present a new method for forecasting systems of multiple interrelated time series. The method learns the forecast models together with discovering leading indicators from within the system that serve as good predictors improving the forecast accuracy and a cluster structure of the predictive tasks around these. The …
We propose a vector auto-regressive (VAR) model with a low-rank constraint on the transition matrix. This new model is well suited to predict high-dimensional series that are highly correlated, or that are driven by a small number of hidden factors. We study estimation, prediction, and rank selection for this model in …
In this paper we investigate the impact of news to predict extreme financial returns using high frequency data. We consider several model specifications differing for the dynamic property of the underlying stochastic process as well as for the innovation process. Since news are essentially qualitative measures, they ar…
This thesis builds a real-time VaR calculation workflow for crypto derivatives.
problem Managing risk in volatile cryptocurrency markets.
method Applied EMWA, GARCH, and HAR models to forecast volatility; used delta-gamma-theta approach and Cornish-Fisher expansion.
result Real-time VaR estimates with millisecond calculation latencies.
Study combines VaR and ES forecasts using MCS to improve risk predictions.
problem Combining VaR and ES forecasts to improve risk predictions under uncertainty.
method Employed Model Confidence Set (MCS) methodology to identify best-performing models and combine their forecasts.
result Proposed combined predictors are robust and pass standard backtests.
QBVAR improves oil price forecasting across quantiles, especially for downside risk.
problem Forecasting oil prices across different quantiles for better risk assessment.
method Quantile Bayesian Vector Autoregression (QBVAR) model.
result QBVAR improves median forecasts by 2-5% and left-tail forecast improvements of 10-25% during crisis episodes.
The analysis of scientific data of increasing size and complexity requires statistical machine learning methods that are both interpretable and predictive. Union of Intersections (UoI), a recently developed framework, is a two-step approach that separates model selection and model estimation. A linear regression algori…
VAR-GPs solve continual learning by updating posteriors sequentially.
problem Catastrophic forgetting in sequential learning tasks.
method Sparse inducing point approximations and auto-regressive variational distribution.
result VAR-GPs prevent catastrophic forgetting and outperform baselines.
Study optimizes stock portfolios using network analysis and forecasting.
problem Optimizing stock portfolios with network analysis and forecasting.
method Constructs dependency networks using VAR and FEVD, applies MST algorithm, and incorporates ARIMA and NNAR forecasts.
result MST-based strategies outperform buy-and-hold benchmarks, achieving higher returns.
Paper proposes MMW distribution for better financial risk modeling.
problem Modeling non-normal stock returns for risk estimation.
method Mixture of mirrored Weibull (MMW) distribution for flexible risk modeling.
result MMW model outperforms Gaussian and t-mixture models in VaR estimation.
Paper proposes a hybrid model for VaR forecasting using SVR, GARCH, and KDE.
problem Inaccurate VaR estimates due to time-varying volatility and distributional characteristics.
method SVR-GARCH-KDE hybrid model combining nonlinear and nonparametric approaches.
result The SVR-GARCH-KDE hybrid outperforms benchmark models in VaR forecasting, especially for longer horizons.
The paper analyzes the joint dynamics of prices and order flow in electronic order books.
problem Understanding the micro-dynamics of asset prices in high-frequency trading environments.
method Double coarse-graining procedure and Principal Component Analysis to extract meaningful information.
result The VAR model captures the stability of liquidity modes and their dynamical evolution.
New method recalibrates VaR for option books, reducing forecast errors.
problem Inaccurate VaR forecasts due to missing operational choices.
method Marking-aware sequential VaR recalibration targeting normalized book-level loss.
result Sequential VaR recalibration improves VaR performance across different markets and options.
This paper estimates VaR for corn and soybean markets using jump processes.
problem Quantifying potential losses in commodity portfolios under market conditions.
method Modeling VaR for a diversified portfolio of corn and soybean positions with standard Brownian motions and jump processes.
result Compared VaR values in markets with and without jumps, providing insights for risk management.
Study uses sentiment analysis to predict implied volatility surface, improving prediction accuracy.
problem Improving prediction accuracy of implied volatility surface.
method Constructed daily high-frequency sentiment data, used VAR method, deep learning (BERT, LSTM), FFT, EMD for sentiment decomposition.
result High-frequency sentiment correlates with ATM options' implied volatility, low-frequency with DOTM options.
Study reveals dynamic linkage between Peanut and Soybean Oil futures markets.
problem Exploring interdependence between Peanut and other agricultural commodities in Chinese futures market.
method Constructed multivariate linear regression models and used VAR and DCC-EGARCH models for dynamic relationships. Applied MLP, CNN, and LSTM neural networks for price prediction.
result Significant dynamic linkage between Peanut and Soybean Oil futures markets through DCC-EGARCH, limited influence from other futures markets through VAR model.
The vector autoregressive (VAR) model is a powerful tool in modeling complex time series and has been exploited in many fields. However, fitting high dimensional VAR model poses some unique challenges: On one hand, the dimensionality, caused by modeling a large number of time series and higher order autoregressive proc…
The paper proposes a new portfolio optimization model that includes VaR risk measure.
problem Computational hardness of portfolio optimization models with VaR as a risk measure.
method Formulated as a Mixed-Integer Quadratic Programming (MIQP) problem, the model minimizes variance with constraints on expected return and VaR.
result The proposed Mean-Variance-VaR portfolios outperform traditional Mean-Variance and Mean-VaR portfolios in out-of-sample performance.
Study improves dividend discount model using VAR process.
problem Improving dividend discount models for better predictions.
method Introduced a Gordon growth model based on Vector Autoregressive Process (VAR).
result Two Propositions related to the new model.
This paper compares VaR estimation methods under tail misspecification, finding importance sampling underestimates VaR.
problem Tail misspecification in VaR estimation.
method Importance sampling and moment-based VaR bracketing.
result Importance sampling underestimates VaR under heavy-tailed returns, while moment-based methods are robust.
Study on forecasting methods and their causal implications.
problem Understanding the difference between statistical and causal risks in forecasting models.
method Introduce causal learning theory for forecasting, obtain uniform convergence bounds for VAR models.
result First theoretical guarantees for causal generalization in time-series forecasting.
Paper proposes a new sparsity scheme for high-dimensional VAR models.
problem Estimation of high-dimensional VAR models with sparsity assumptions.
method Regularized estimation procedures for sparse VAR models.
result Threholding extends consistency properties of regularized estimators.
This thesis examines the accuracy of scaling VaR estimates for longer holding periods.
problem The accuracy of VaR estimates for longer holding periods using the square root of time rule.
method Examined VaR scaling for longer holding periods using empirical analysis.
result Scaling can provide good estimates of VaR but may lead to significant losses over time.
Linear attention in Transformers can be interpreted as dynamic VAR models.
problem Misalignment between Transformers and autoregressive forecasting objectives.
method Interpreting linear attention as VAR, rearranging MLP, attention, and flow.
result SAMoVAR improves performance, interpretability, and efficiency.
Paper proposes GAS-ALD model for financial risk prediction.
problem Skewed distribution of financial return data.
method Generalized autoregressive score (GAS) framework with asymmetric Laplace distribution (ALD).
result GAS-ALD model predicts VaR and ES more accurately than traditional models.
Investment strategy for DC pension plan with inflation risk and tail VaR constraint.
problem Maximizing terminal wealth for pension member with tail VaR constraint.
method Lagrange method and quantile optimization techniques.
result Optimal investment strategy and output in closed-form derived.
VaR-CPO optimizes VaR-constrained RL problems with conservative policy updates.
problem Optimizing VaR-constrained reinforcement learning problems.
method Combines Cantelli's inequality and trust-region framework for efficient and conservative optimization.
result Achieves zero constraint violations during training in feasible environments.
Pricing and hedging rainbow options using Bayesian MS-VAR process.
problem Pricing and hedging rainbow options under varying economic conditions.
method Bayesian Markov-Switching Vector Autoregressive (MS-VAR) process to model regime-switching economic variables.
result Model provides a simpler and more economic variable-dependent approach for rainbow options pricing and hedging.
Paper investigates Lambda Value-at-Risk under ambiguity and risk sharing.
problem Investigates Lambda Value-at-Risk under ambiguity and risk sharing.
method Establishes equivalence of robust ΛVaR and traditional ΛVaR under ambiguity sets, analyzes properties, derives explicit formulas, and explores risk sharing. result Unified and extended the concept of Value-at-Risk under ambiguity, derived explicit formulas for specific ambiguity sets, and explored risk sharing.
A dynamic Boltzmann machine (DyBM) has been proposed as a model of a spiking neural network, and its learning rule of maximizing the log-likelihood of given time-series has been shown to exhibit key properties of spike-timing dependent plasticity (STDP), which had been postulated and experimentally confirmed in the fie…