ATAnastasija Tetereva

Anastasija Tetereva

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Anastasija Tetereva has a total of 13 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Can Machines Learn Electricity Hedging Regimes? State-Dependent Minimum-Variance Hedging in German Power Futures

    by Anastasija Tetereva, Evghenii Surujiu, Marts Vecvanags, Lovro Andrić, Linuka Koththigoda

    Wholesale power prices are volatile and prone to extreme spikes, and the minimum variance hedge ratio linking spot and futures returns is not constant. We propose a state-dependent hedging framework that keeps the minimum-variance hedge as a transparent regression of spot on futures price changes but lets a feed-forward neural network weight each past week by its similarity to prevailing market conditions, so the effective estimation sample adapts to the current regime while the hedge ratio itself remains a single interpretable coefficient. The similarity metric is learned from an economic state vector of fuel and carbon prices, renewable generation, load, weather, and the spot-futures basis. Using weekly German baseload power futures from 2019 to 2026, we evaluate the hedge against a comprehensive set of benchmarks and find that it delivers the highest full-period point-estimate hedge effectiveness while materially improving downside-risk performance relative to expanding OLS. The improvement is largest where hedgers need it most: downside-semivariance hedge effectiveness rises from 0.499 to 0.671. The gains coincide with high fuel-price and wide-basis stress states, while the estimated weights show selective concentration on stressed historical episodes. Economically, the enhanced hedge improves estimated 5% conditional value at risk by roughly 10 EUR/MWh, corresponding to a reduction in tail-loss magnitude. For a notional 50 MW baseload buyer, this implies an indicative weekly capital-at-risk reduction with a point estimate of about EUR 85 thousand and a 90% bootstrap interval of roughly EUR 2–201 thousand.

  • Open Access

    Nelson-Siegel Autoencoders for Global Yield Curve Forecasting

    by Aaron Popa, Maria Grith, Anastasija Tetereva

    We study global yield curve dynamics using an economics-informed autoencoder framework that captures the hierarchical structure of international bond markets. We propose a Global Nelson-Siegel Autoencoder anchored in the Nelson-Siegel basis that decomposes a panel of yield curves into global and country-specific components. For forecasting, we introduce a Dynamic Global Nelson-Siegel Autoencoder that jointly learns latent factors and their linear time-series dynamics. Using monthly government bond yields for the United States, Germany, the United Kingdom, and Japan from 1995 to 2023, we show that global factors explain a large and time-varying share of yield variation. Global models deliver robust out-of-sample forecasting gains for the United States and the United Kingdom, particularly during periods of heightened global uncertainty. Forecasting improvements increase with the horizon, consistent with global factors being especially informative for longer-term risk compensation.

  • Open Access

    Asset Pricing Factors with Conditionally Sorted Characteristics

    by Gustavo Freire, Rasmus Lönn, Anastasija Tetereva

    We propose a new approach for constructing characteristics-based factor portfolios. We allocate portfolio weights proportional to the conditional rank of each characteristic given all other characteristics, which generalizes quantile sorts to better isolate the premium associated with each factor. We model this conditional rank in a simple and data-driven way using copulas. We apply our method to the five Fama-French factor portfolios. Relative to the original construction, our conditional-rank factors increase the maximum attainable Sharpe ratio from 1.04 to 1.80 and decrease by half the number of anomalies with significant alphas in the cross-section of stock returns.

  • Open Access

    Forecast Combination for Tail Risk with Regulator-Aware Decision Trees

    by Anastasija Tetereva, Ekaterina Kazak

    In this paper, we address the challenge of combining forecasts for Value-at-Risk (VaR) and Expected Shortfall (ES) by employing nonparametric weights which are functions of macro-financial state variables. We develop a tailor-made, tree-based machine learning procedure in which optimal combination weights remain constant within economic regimes but vary across them. The continuously varying regimes themselves are identified in a fully data-driven manner. To ensure consistency with supervisory practice, we align the tree construction process with liquidity-horizonspecific risk factors defined by the Basel Committee on Banking Supervision. We further reduce the influence of correlated covariates by proposing a methodology of cluster-wise random variable draws when building each tree. In an application to daily returns of 24 large-cap U.S. stocks from 2000 to 2023, the proposed marketdriven forecast combination delivers significantly lower out-of-sample VaR-ES losses than the standard relative-score benchmark for up to 22 out of 24 assets across multiple liquidity horizons, while also improving coverage properties and economic interpretability.

  • Open Access

    Forecasting Realized Correlations: A MIDAS Approach

    by Alexander Kostrov, Anastasija Tetereva

    Mixed data sampling (MIDAS) regression has received much attention in relation to modeling financial time series due to its flexibility. Previous work has mainly focused on forecasting of realized volatilities and has rarely been used to predict realized correlations. This paper considers a MIDAS approach to forecast realized correlation matrices. A MIDAS model is estimated via nonlinear least squares (NLS) using an analytical gradient-based optimization. Based on the model confidence set (MCS) procedure we discover that the introduced approach is superior compared to the established heterogeneous autoregressive (HAR) model in terms of out-of-sample forecasting accuracy. This preeminence is due to the flexible data-driven origin of the MIDAS model. The latter results in higher economic value with regard to portfolio management applications. The improvement is considerable for longer forecasting horizons both in calm times and during the periods of market turbulence.

  • Open Access

    The Realized Hierarchical Archimedean Copula in Risk Modelling

    by Ostap Okhrin, Anastasija Tetereva

    This paper introduces the concept of the realized hierarchical Archimedean copula (rHAC). The proposed approach inherits the ability of the copula to capture the dependencies among financial time series, and combines it with additional information contained in high-frequency data. The considered model does not suffer from the curse of dimensionality, and is able to accurately predict high-dimensional distributions. This flexibility is obtained by using a hierarchical structure in the copula. The time variability of the model is provided by daily forecasts of the realized correlation matrix, which is used to estimate the structure and the parameters of the rHAC. Extensive simulation studies show the validity of the estimator based on this realized correlation matrix, and its performance, in comparison to the benchmark models. The application of the estimator to one-day-ahead Value at Risk (VaR) prediction using high-frequency data exhibits good forecasting properties for a multivariate portfolio.


  • Open Access

    Sentiment Spillover Effects for US and European Companies

    by Francesco Audrino, Anastasija Tetereva

    The fast-growing literature on the news and social media analysis provide empirical evidence that the financial markets are often driven by information rather than facts. However, the direct effects of sentiments on the returns are of main interest. In this paper, we propose to study the cross-industry influence of the news for a set of US and European stocks. The graphical Granger causality of the news sentiments - excess return networks is estimated by applying the adaptive Lasso procedure. We introduce two characteristics to measure the influence of the news coming from each sector and analyze their dynamics for a period of 10 years ranging from 2005 to 2014. The results obtained provide insight into the news spillover effects among the industries and the importance of sentiments related to certain sectors during periods of financial instability.