{
  "generated_at_utc": "2026-03-12 11:05:46",
  "days": 7,
  "top": 5,
  "min_score": 35,
  "include_seen": false,
  "allow_undated": false,
  "selected_count": 5,
  "dives": [
    {
      "paper": {
        "id": "2603.05917v1",
        "source": "arxiv",
        "title": "Stock Market Prediction Using Node Transformer Architecture Integrated with BERT Sentiment Analysis",
        "abstract": "Stock market prediction presents considerable challenges for investors, financial institutions, and policymakers operating in complex market environments characterized by noise, non-stationarity, and behavioral dynamics. Traditional forecasting methods often fail to capture the intricate patterns and cross-sectional dependencies inherent in financial markets. This paper presents an integrated framework combining a node transformer architecture with BERT-based sentiment analysis for stock price forecasting. The proposed model represents the stock market as a graph structure where individual stocks form nodes and edges capture relationships including sectoral affiliations, correlated price movements, and supply chain connections. A fine-tuned BERT model extracts sentiment from social media posts and combines it with quantitative market features through attention-based fusion. The node transformer processes historical market data while capturing both temporal evolution and cross-sectional dependencies among stocks. Experiments on 20 S&P 500 stocks spanning January 1982 to March 2025 demonstrate that the integrated model achieves a mean absolute percentage error (MAPE) of 0.80% for one-day-ahead predictions, compared to 1.20% for ARIMA and 1.00% for LSTM. Sentiment analysis reduces prediction error by 10% overall and 25% during earnings announcements, while graph-based modeling contributes an additional 15% improvement by capturing inter-stock dependencies. Directional accuracy reaches 65% for one-day forecasts. Statistical validation through paired t-tests confirms these improvements (p < 0.05 for all comparisons). The model maintains MAPE below 1.5% during high-volatility periods where baseline models exceed 2%.",
        "authors": [
          "Mohammad Al Ridhawi",
          "Mahtab Haj Ali",
          "Hussein Al Osman"
        ],
        "date": "2026-03-06",
        "categories": [
          "cs.LG",
          "cs.AI",
          "q-fin.ST"
        ],
        "url": "https://arxiv.org/abs/2603.05917v1",
        "pdf": "https://arxiv.org/pdf/2603.05917v1",
        "relevance_score": 76,
        "high_keywords": [
          "s&p 500",
          "earnings",
          "announcement"
        ],
        "medium_keywords": [
          "sentiment",
          "lstm",
          "transformer"
        ],
        "low_keywords": [],
        "actionable": true,
        "findings": [],
        "methods": [
          "LSTM",
          "Transformer model",
          "Cross-sectional analysis"
        ],
        "backtestable": false
      },
      "analysis": {
        "paper": {
          "id": "2603.05917v1",
          "title": "Stock Market Prediction Using Node Transformer Architecture Integrated with BERT Sentiment Analysis",
          "abstract": "Stock market prediction presents considerable challenges for investors, financial institutions, and policymakers operating in complex market environments characterized by noise, non-stationarity, and behavioral dynamics. Traditional forecasting methods often fail to capture the intricate patterns and cross-sectional dependencies inherent in financial markets. This paper presents an integrated framework combining a node transformer architecture with BERT-based sentiment analysis for stock price forecasting. The proposed model represents the stock market as a graph structure where individual stocks form nodes and edges capture relationships including sectoral affiliations, correlated price movements, and supply chain connections. A fine-tuned BERT model extracts sentiment from social media posts and combines it with quantitative market features through attention-based fusion. The node transformer processes historical market data while capturing both temporal evolution and cross-sectional dependencies among stocks. Experiments on 20 S&P 500 stocks spanning January 1982 to March 2025 demonstrate that the integrated model achieves a mean absolute percentage error (MAPE) of 0.80% for one-day-ahead predictions, compared to 1.20% for ARIMA and 1.00% for LSTM. Sentiment analysis reduces prediction error by 10% overall and 25% during earnings announcements, while graph-based modeling contributes an additional 15% improvement by capturing inter-stock dependencies. Directional accuracy reaches 65% for one-day forecasts. Statistical validation through paired t-tests confirms these improvements (p < 0.05 for all comparisons). The model maintains MAPE below 1.5% during high-volatility periods where baseline models exceed 2%.",
          "authors": [
            "Mohammad Al Ridhawi",
            "Mahtab Haj Ali",
            "Hussein Al Osman"
          ],
          "date": "2026-03-06",
          "url": "https://arxiv.org/abs/2603.05917v1",
          "pdf": "https://arxiv.org/pdf/2603.05917v1"
        },
        "analysis_date": "2026-03-12 04:05",
        "asset_classes": [
          "equity"
        ],
        "time_horizons": [
          "daily"
        ],
        "data_requirements": [
          "price_data",
          "fundamental",
          "sentiment"
        ],
        "reported_metrics": {},
        "es_applicability": 4,
        "replicable_with_our_data": false
      },
      "priority_score": 87.0,
      "tradeable_notes": [
        "General research candidate; map methodology to existing ES feature pipeline.",
        "May require additional data source mapping before backtest implementation."
      ]
    },
    {
      "paper": {
        "id": "2603.09164v1",
        "source": "arxiv",
        "title": "Slippage-at-Risk (SaR): A Forward-Looking Liquidity Risk Framework for Perpetual Futures Exchanges",
        "abstract": "We introduce $\\textbf{Slippage-at-Risk (SaR)}$, a quantitative framework for measuring liquidity risk in perpetual futures exchanges. Unlike backward-looking metrics such as Value-at-Risk computed on historical returns or realized deficit distributions, SaR provides a \\emph{forward-looking} assessment of liquidation execution risk derived from current order book microstructure. The framework comprises three complementary metrics: $SaR(\u03b1)$, the cross-sectional slippage quantile; $ESaR(\u03b1)$, the expected slippage in the distributional tail; and $TSaR(\u03b1)$, the aggregate dollar-denominated tail slippage. We extend the base framework with a \\emph{concentration adjustment} that penalizes fragile liquidity structures where a small number of market makers dominate quote provision. Drawing on recent work by Chitra et al. (2025) on autodeleveraging mechanisms and insurance fund optimization, we establish a direct mapping from SaR metrics to optimal capital requirements. Empirical analysis using Hyperliquid order book data, including the October 10, 2025 liquidation cascade, demonstrates SaR's predictive validity as a leading indicator of systemic stress. We conclude with practical implementation guidance and discuss philosophical implications for risk management in decentralized financial systems.",
        "authors": [
          "Otar Sepper"
        ],
        "date": "2026-03-10",
        "categories": [
          "q-fin.RM"
        ],
        "url": "https://arxiv.org/abs/2603.09164v1",
        "pdf": "https://arxiv.org/pdf/2603.09164v1",
        "relevance_score": 48,
        "high_keywords": [
          "market maker",
          "microstructure",
          "liquidity",
          "market maker"
        ],
        "medium_keywords": [],
        "low_keywords": [
          "defi",
          "insurance"
        ],
        "actionable": true,
        "findings": [],
        "methods": [
          "Cross-sectional analysis"
        ],
        "backtestable": true
      },
      "analysis": {
        "paper": {
          "id": "2603.09164v1",
          "title": "Slippage-at-Risk (SaR): A Forward-Looking Liquidity Risk Framework for Perpetual Futures Exchanges",
          "abstract": "We introduce $\\textbf{Slippage-at-Risk (SaR)}$, a quantitative framework for measuring liquidity risk in perpetual futures exchanges. Unlike backward-looking metrics such as Value-at-Risk computed on historical returns or realized deficit distributions, SaR provides a \\emph{forward-looking} assessment of liquidation execution risk derived from current order book microstructure. The framework comprises three complementary metrics: $SaR(\u03b1)$, the cross-sectional slippage quantile; $ESaR(\u03b1)$, the expected slippage in the distributional tail; and $TSaR(\u03b1)$, the aggregate dollar-denominated tail slippage. We extend the base framework with a \\emph{concentration adjustment} that penalizes fragile liquidity structures where a small number of market makers dominate quote provision. Drawing on recent work by Chitra et al. (2025) on autodeleveraging mechanisms and insurance fund optimization, we establish a direct mapping from SaR metrics to optimal capital requirements. Empirical analysis using Hyperliquid order book data, including the October 10, 2025 liquidation cascade, demonstrates SaR's predictive validity as a leading indicator of systemic stress. We conclude with practical implementation guidance and discuss philosophical implications for risk management in decentralized financial systems.",
          "authors": [
            "Otar Sepper"
          ],
          "date": "2026-03-10",
          "url": "https://arxiv.org/abs/2603.09164v1",
          "pdf": "https://arxiv.org/pdf/2603.09164v1"
        },
        "analysis_date": "2026-03-12 04:05",
        "asset_classes": [
          "options",
          "futures"
        ],
        "time_horizons": [
          "unspecified"
        ],
        "data_requirements": [
          "price_data",
          "order_flow"
        ],
        "reported_metrics": {},
        "es_applicability": 2,
        "replicable_with_our_data": true
      },
      "priority_score": 73.0,
      "tradeable_notes": [
        "Candidate for intraday execution filter or 30m/1h tactical signal test.",
        "Replicable with current local data stack (price/options/order flow)."
      ]
    },
    {
      "paper": {
        "id": "2603.09006v1",
        "source": "arxiv",
        "title": "Spectral Portfolio Theory: From SGD Weight Matrices to Wealth Dynamics",
        "abstract": "We develop spectral portfolio theory by establishing a direct identification: neural network weight matrices trained on stochastic processes are portfolio allocation matrices, and their spectral structure encodes factor decompositions and wealth concentration patterns. The three forces governing stochastic gradient descent (SGD) -- gradient signal, dimensional regularisation, and eigenvalue repulsion -- translate directly into portfolio dynamics: smart money, survival constraint, and endogenous diversification. The spectral properties of SGD weight matrices transition from Marchenko-Pastur statistics (additive regime, short horizon) to inverse-Wishart via the free log-normal (multiplicative regime, long horizon), mirroring the transition from daily returns to long-run wealth compounding. We unify the cross-sectional wealth dynamics of Bouchaud and Mezard (2000), the within-portfolio dynamics of Olsen et al. (2025), and the scalar Fokker-Planck framework via a common spectral foundation. A central result is the Spectral Invariance Theorem: any isotropic perturbation to the portfolio objective preserves the singular-value distribution up to scale and shift, while anisotropic perturbations produce spectral distortion proportional to their cross-asset variance. We develop applications to portfolio design, wealth inequality measurement, tax policy, and neural network diagnostics. In the tax context, the invariance result recovers and generalises the neutrality conditions of Fr\u00f8seth (2026).",
        "authors": [
          "Anders G Fr\u00f8seth"
        ],
        "date": "2026-03-09",
        "categories": [
          "q-fin.PM",
          "physics.soc-ph"
        ],
        "url": "https://arxiv.org/abs/2603.09006v1",
        "pdf": "https://arxiv.org/pdf/2603.09006v1",
        "relevance_score": 55,
        "high_keywords": [
          "cross-asset",
          "factor",
          "decomposition"
        ],
        "medium_keywords": [
          "neural network",
          "allocation"
        ],
        "low_keywords": [],
        "actionable": true,
        "findings": [],
        "methods": [
          "Neural network",
          "Cross-sectional analysis"
        ],
        "backtestable": false
      },
      "analysis": {
        "paper": {
          "id": "2603.09006v1",
          "title": "Spectral Portfolio Theory: From SGD Weight Matrices to Wealth Dynamics",
          "abstract": "We develop spectral portfolio theory by establishing a direct identification: neural network weight matrices trained on stochastic processes are portfolio allocation matrices, and their spectral structure encodes factor decompositions and wealth concentration patterns. The three forces governing stochastic gradient descent (SGD) -- gradient signal, dimensional regularisation, and eigenvalue repulsion -- translate directly into portfolio dynamics: smart money, survival constraint, and endogenous diversification. The spectral properties of SGD weight matrices transition from Marchenko-Pastur statistics (additive regime, short horizon) to inverse-Wishart via the free log-normal (multiplicative regime, long horizon), mirroring the transition from daily returns to long-run wealth compounding. We unify the cross-sectional wealth dynamics of Bouchaud and Mezard (2000), the within-portfolio dynamics of Olsen et al. (2025), and the scalar Fokker-Planck framework via a common spectral foundation. A central result is the Spectral Invariance Theorem: any isotropic perturbation to the portfolio objective preserves the singular-value distribution up to scale and shift, while anisotropic perturbations produce spectral distortion proportional to their cross-asset variance. We develop applications to portfolio design, wealth inequality measurement, tax policy, and neural network diagnostics. In the tax context, the invariance result recovers and generalises the neutrality conditions of Fr\u00f8seth (2026).",
          "authors": [
            "Anders G Fr\u00f8seth"
          ],
          "date": "2026-03-09",
          "url": "https://arxiv.org/abs/2603.09006v1",
          "pdf": "https://arxiv.org/pdf/2603.09006v1"
        },
        "analysis_date": "2026-03-12 04:05",
        "asset_classes": [],
        "time_horizons": [
          "daily"
        ],
        "data_requirements": [
          "price_data"
        ],
        "reported_metrics": {},
        "es_applicability": 2,
        "replicable_with_our_data": true
      },
      "priority_score": 68.0,
      "tradeable_notes": [
        "General research candidate; map methodology to existing ES feature pipeline.",
        "Replicable with current local data stack (price/options/order flow)."
      ]
    },
    {
      "paper": {
        "id": "2603.05862v1",
        "source": "arxiv",
        "title": "Impact of arbitrage between leveraged ETF and futures on market liquidity during market crash",
        "abstract": "Leveraged ETFs (L-ETFs) are exchange-traded funds that achieve price movements several times greater than an index by holding index-linked futures such as Nikkei Stock Average Index futures. It is known that when the price of an L-ETF falls, the L-ETF uses the liquidity of futures to limit the decline through arbitrage trading. Conversely, when the price of a futures contract falls, the futures contract uses the liquidity of the L-ETF to limit its decline. However, the impact of arbitrage trading on the liquidity of these markets has been little studied. Therefore, the present study used artificial market simulations to investigate how the liquidity (Volume, SellDepth, BuyDepth, Tightness) of both markets changes when prices plummet in either (i.e., the L-ETF or futures market), depending on the presence or absence of arbitrage trading. As a result, it was found that when erroneous orders occur in the L-ETF market, the existence of arbitrage trading causes liquidity to be supplied from the futures market to the L-ETF market in terms of SellDepth and Tightness. When erroneous orders occur in the futures market, the existence of arbitrage trading causes liquidity to be supplied from the L-ETF market to the futures market in terms of SellDepth and Tightness, and liquidity to be supplied from the futures market to the L-ETF market in terms of Volume. We also analyzed the internal market mechanisms that led to these results.",
        "authors": [
          "Ryuki Hayase",
          "Takanobu Mizuta",
          "Isao Yagi"
        ],
        "date": "2026-03-06",
        "categories": [
          "q-fin.CP",
          "cs.MA"
        ],
        "url": "https://arxiv.org/abs/2603.05862v1",
        "pdf": "https://arxiv.org/pdf/2603.05862v1",
        "relevance_score": 50,
        "high_keywords": [
          "index futures",
          "liquidity",
          "futures market"
        ],
        "medium_keywords": [
          "etf"
        ],
        "low_keywords": [],
        "actionable": true,
        "findings": [],
        "methods": [],
        "backtestable": false
      },
      "analysis": {
        "paper": {
          "id": "2603.05862v1",
          "title": "Impact of arbitrage between leveraged ETF and futures on market liquidity during market crash",
          "abstract": "Leveraged ETFs (L-ETFs) are exchange-traded funds that achieve price movements several times greater than an index by holding index-linked futures such as Nikkei Stock Average Index futures. It is known that when the price of an L-ETF falls, the L-ETF uses the liquidity of futures to limit the decline through arbitrage trading. Conversely, when the price of a futures contract falls, the futures contract uses the liquidity of the L-ETF to limit its decline. However, the impact of arbitrage trading on the liquidity of these markets has been little studied. Therefore, the present study used artificial market simulations to investigate how the liquidity (Volume, SellDepth, BuyDepth, Tightness) of both markets changes when prices plummet in either (i.e., the L-ETF or futures market), depending on the presence or absence of arbitrage trading. As a result, it was found that when erroneous orders occur in the L-ETF market, the existence of arbitrage trading causes liquidity to be supplied from the futures market to the L-ETF market in terms of SellDepth and Tightness. When erroneous orders occur in the futures market, the existence of arbitrage trading causes liquidity to be supplied from the L-ETF market to the futures market in terms of SellDepth and Tightness, and liquidity to be supplied from the futures market to the L-ETF market in terms of Volume. We also analyzed the internal market mechanisms that led to these results.",
          "authors": [
            "Ryuki Hayase",
            "Takanobu Mizuta",
            "Isao Yagi"
          ],
          "date": "2026-03-06",
          "url": "https://arxiv.org/abs/2603.05862v1",
          "pdf": "https://arxiv.org/pdf/2603.05862v1"
        },
        "analysis_date": "2026-03-12 04:05",
        "asset_classes": [
          "equity",
          "futures"
        ],
        "time_horizons": [
          "unspecified"
        ],
        "data_requirements": [
          "price_data",
          "order_flow"
        ],
        "reported_metrics": {},
        "es_applicability": 3,
        "replicable_with_our_data": true
      },
      "priority_score": 61.0,
      "tradeable_notes": [
        "General research candidate; map methodology to existing ES feature pipeline.",
        "Replicable with current local data stack (price/options/order flow)."
      ]
    },
    {
      "paper": {
        "id": "2603.07616v1",
        "source": "arxiv",
        "title": "SABR Type Libor (Forward) Market Model (SABR/LMM) with time-dependent skew and smile",
        "abstract": "Volatility Skew and Smile of Interest Rate products (Swaption and Caplet) are represented by SABR (Stochastic Alpha Beta Rho model). So, the Interest Rate derivatives model for pricing the callable exotic swaps should be comparable to the SABR volatility surface. In the interest rate derivatives models, Libor Market Model (LMM) (in a post-Libor world, Forward Market Model (FMM)) is one of the most popular models used in the market. So, there are many attempts to develop LMMs that are comparable to the SABR surface. It is called SABR/LMM. There are many references for SABR/LMM, but most of them only treat SABR/LMM, which is not flexible enough to be used practically in global banks. The purpose of this paper is to provide a comprehensive definition of SABR/LMM and a complete description of how it is to be implemented.",
        "authors": [
          "Osamu Tsuchiya"
        ],
        "date": "2026-03-08",
        "categories": [
          "q-fin.MF",
          "q-fin.PR"
        ],
        "url": "https://arxiv.org/abs/2603.07616v1",
        "pdf": "https://arxiv.org/pdf/2603.07616v1",
        "relevance_score": 40,
        "high_keywords": [
          "volatility surface",
          "skew",
          "volatility surface"
        ],
        "medium_keywords": [
          "alpha"
        ],
        "low_keywords": [
          "defi"
        ],
        "actionable": true,
        "findings": [],
        "methods": [],
        "backtestable": false
      },
      "analysis": {
        "paper": {
          "id": "2603.07616v1",
          "title": "SABR Type Libor (Forward) Market Model (SABR/LMM) with time-dependent skew and smile",
          "abstract": "Volatility Skew and Smile of Interest Rate products (Swaption and Caplet) are represented by SABR (Stochastic Alpha Beta Rho model). So, the Interest Rate derivatives model for pricing the callable exotic swaps should be comparable to the SABR volatility surface. In the interest rate derivatives models, Libor Market Model (LMM) (in a post-Libor world, Forward Market Model (FMM)) is one of the most popular models used in the market. So, there are many attempts to develop LMMs that are comparable to the SABR surface. It is called SABR/LMM. There are many references for SABR/LMM, but most of them only treat SABR/LMM, which is not flexible enough to be used practically in global banks. The purpose of this paper is to provide a comprehensive definition of SABR/LMM and a complete description of how it is to be implemented.",
          "authors": [
            "Osamu Tsuchiya"
          ],
          "date": "2026-03-08",
          "url": "https://arxiv.org/abs/2603.07616v1",
          "pdf": "https://arxiv.org/pdf/2603.07616v1"
        },
        "analysis_date": "2026-03-12 04:05",
        "asset_classes": [
          "options",
          "fixed_income"
        ],
        "time_horizons": [
          "unspecified"
        ],
        "data_requirements": [],
        "reported_metrics": {},
        "es_applicability": 2,
        "replicable_with_our_data": false
      },
      "priority_score": 51.0,
      "tradeable_notes": [
        "Candidate for volatility-context overlay (position sizing / regime filter).",
        "May require additional data source mapping before backtest implementation."
      ]
    }
  ]
}