We introduce M2VN: Multi-Modal Volatility Network, a novel deep learning-based framework for financial volatility forecasting that unifies time series features with unstructured news data. M2VN leverages the representational power of deep neural networks to address two key challenges in this domain: (i) aligning and fusing heterogeneous data modalities, numerical financial data and textual information, and (ii) mitigating look-ahead bias that can undermine the validity of financial models. To achieve this, M2VN combines open-source market features with news embeddings generated by Time Machine GPT, a recently introduced point-in-time LLM, ensuring temporal integrity. An auxiliary alignment loss is introduced to enhance the integration of structured and unstructured data within the deep learning architecture. Extensive experiments demonstrate that M2VN consistently outperforms existing baselines, underscoring its practical value for risk management and financial decision-making in dynamic markets.

Abstract--Financial news sentiment analysis is crucial for anticipating market movements. With the rise of AI techniques such as Large Language Models (LLMs), which demonstrate strong text understanding capabilities, there has been renewed interest in enhancing these systems. Existing methods, however, often struggle to capture the complex economic context of news and lack transparent reasoning, which undermines their reliability. We propose Analogy-Driven Financial Chain-of-Thought (AD-FCoT), a prompting framework that integrates analogical reasoning with chain-of-thought (CoT) prompting for sentiment prediction on historical financial news. AD-FCoT guides LLMs to draw parallels between new events and relevant historical scenarios with known outcomes, embedding these analogies into a structured, step-by-step reasoning chain. T o our knowledge, this is among the first approaches to explicitly combine analogical examples with CoT reasoning in finance. Operating purely through prompting, AD-FCoT requires no additional training data or fine-tuning and leverages the model's internal financial knowledge to generate rationales that mirror human analytical reasoning. Experiments on thousands of news articles show that AD-FCoT outperforms strong baselines in sentiment classification accuracy and achieves substantially higher correlation with market returns. Its generated explanations also align with domain expertise, providing interpretable insights suitable for real-world financial analysis.

Emerging techniques in computer science make it possible to "brain scan" large language models (LLMs), identify the plain-English concepts that guide their reasoning, and steer them while holding other factors constant. We show that this approach can map LLM-generated economic forecasts to concepts such as sentiment, technical analysis, and timing, and compute their relative importance without reducing performance. We also show that models can be steered to be more or less risk-averse, optimistic, or pessimistic, which allows researchers to correct or simulate biases. The method is transparent, lightweight, and replicable for empirical research in the social sciences.

Large language models (LLMs), including ChatGPT, can extract profitable trading signals from the sentiment in news text. However, backtesting such strategies poses a challenge because LLMs are trained on many years of data, and backtesting produces biased results if the training and backtesting periods overlap. This bias can take two forms: a look-ahead bias, in which the LLM may have specific knowledge of the stock returns that followed a news article, and a distraction effect, in which general knowledge of the companies named interferes with the measurement of a text's sentiment. We investigate these sources of bias through trading strategies driven by the sentiment of financial news headlines. We compare trading performance based on the original headlines with de-biased strategies in which we remove the relevant company's identifiers from the text. In-sample (within the LLM training window), we find, surprisingly, that the anonymized headlines outperform, indicating that the distraction effect has a greater impact than look-ahead bias. This tendency is particularly strong for larger companies--companies about which we expect an LLM to have greater general knowledge. Out-of-sample, look-ahead bias is not a concern but distraction remains possible. Our proposed anonymization procedure is therefore potentially useful in out-of-sample implementation, as well as for de-biased backtesting.

In Time Series Forecasting, deep learning neural networks have just lately outperformed conventional techniques and have done so by a smaller margin than in image and language processing. The days of creating a model specifically for a single time series, whether it be multivariate or univariate, are long gone. Nowadays, time series could be multivariate, have various distributions, and include more exploratory factors. The usual suspects, such as missing data, trends, seasonality, volatility, drift, and rare events, should also not be overlooked. A straightforward target variable prediction is frequently insufficient.

Promising results were obtained with the gradient boosting classifier. Predicting the success of a business venture has always been a struggle for both practitioners and researchers. However, thanks to companies that aggregate data about other firms, it has become possible to create and validate predictive models based on an unprecedented amount of real-world examples. In this study, we use data obtained from one of the largest platforms integrating business information – Crunchbase. Our final training set consisted of 213 171 companies.

Data scientists are in short supply, and Automatic Machine Learning (AutoML) promises to alleviate this problem. H2O Driverless AI employs the techniques of expert data scientists in an easy to use application that helps scale your data science efforts. It lets "everyone develop trusted machine learning models." A Chief Data Scientist and four Columbia University students with varying levels of experience put the technology to the test on a real-word problem predicting stock price movements (a challenging task!). The data was 5 years of stocks in the Russell 1000 index on a monthly frequency.

The least square Monte Carlo (LSM) algorithm proposed by Longstaff and Schwartz [2001] is widely used for pricing American options. The LSM estimator contains undesirable look-ahead bias, and the conventional technique of removing it necessitates doubling simulations. We present the leave-one-out LSM (LOOLSM) algorithm for efficiently eliminating look-ahead bias. We validate the method with several option examples, including the multi-asset cases that the LSM algorithm significantly overvalues. We also obtain the convergence rates of look-ahead bias by measuring it using the LOOLSM method. The analysis and computational evidence support our findings.

Traditionally, Back testing is a data based approach to decision making. Backtesting offers Research analysts, traders, and investors a way to evaluate and optimize their trading strategies and model portfolios before implementing them. This is done by using historical data, backtest the model to see whether it would have worked in the past. By comparing the predicted results of the model against the actual historical results, backtesting can determine whether the model has predictive value. This is where we, in essence, put your trading strategies and model portfolios into a time machine (i.e.