Sanjiv Das’s research while affiliated with Santa Clara University and other places

We study how globalization impacts systemic risk in emerging markets. We extend a large literature on systemic risk in the US, Europe, and other developed countries to emerging markets, which are relatively under-researched. Our findings are based on a large-scale empirical examination of systemic risk among 1048 financial institutions in a sample of 23 emerging markets, broken down into 5 regions, along with 369 U.S. financial institutions. Using an additively decomposable systemic risk score that combines banking system interconnectedness with default probabilities, systemic risk is quantified for each region, across time. The empirical analyses suggest that emerging markets’ systemic risk is heterogeneous across regions, is strongly dependent on the interconnectedness of the banking system within each region, and drives the level of default risk in each region, while the regions are compartmentalized away from each other and insulated from the United States. The systemic risk score may be used as a policy variable in each emerging market region to manage the credit cycle. Our evidence is consistent with the notion that globalization engenders financial stability and does not lead to large systemic risk spillovers across emerging market regions.

We develop a methodology for converting card index archival records into usable data frames for statistical and textual analyses. Leveraging machine learning and natural-language processing tools from Amazon Web Services (AWS), we overcome hurdles associated with character recognition, inconsistent data reporting, column misalignment, and irregular naming. In this article, we detail the step-by-step conversion process and discuss remedies for common problems and edge cases, using historical records from the Reconstruction Finance Corporation.

Data-driven methods that detect anomalies in times series data are ubiquitous in practice, but they are in general unable to provide helpful explanations for the predictions they make. In this work we propose a model-agnostic algorithm that generates counterfactual ensemble explanations for time series anomaly detection models. Our method generates a set of diverse counterfactual examples, i.e, multiple perturbed versions of the original time series that are not considered anomalous by the detection model. Since the magnitude of the perturbations is limited, these counterfactuals represent an ensemble of inputs similar to the original time series that the model would deem normal. Our algorithm is applicable to any differentiable anomaly detection model. We investigate the value of our method on univariate and multivariate real-world datasets and two deep-learning-based anomaly detection models, under several explainability criteria previously proposed in other data domains such as Validity, Plausibility, Closeness and Diversity. We show that our algorithm can produce ensembles of counterfactual examples that satisfy these criteria and thanks to a novel type of visualisation, can convey a richer interpretation of a model's internal mechanism than existing methods. Moreover, we design a sparse variant of our method to improve the interpretability of counterfactual explanations for high-dimensional time series anomalies. In this setting, our explanation is localised on only a few dimensions and can therefore be communicated more efficiently to the model's user.

The large size and complex decision mechanisms of state-of-the-art text classifiers make it difficult for humans to understand their predictions, leading to a potential lack of trust by the users. These issues have led to the adoption of methods like SHAP and Integrated Gradients to explain classification decisions by assigning importance scores to input tokens. However, prior work, using different randomization tests, has shown that interpretations generated by these methods may not be robust. For instance, models making the same predictions on the test set may still lead to different feature importance rankings. In order to address the lack of robustness of token-based interpretability, we explore explanations at higher semantic levels like sentences. We use computational metrics and human subject studies to compare the quality of sentence-based interpretations against token-based ones. Our experiments show that higher-level feature attributions offer several advantages: 1) they are more robust as measured by the randomization tests, 2) they lead to lower variability when using approximation-based methods like SHAP, and 3) they are more intelligible to humans in situations where the linguistic coherence resides at a higher granularity level. Based on these findings, we show that token-based interpretability, while being a convenient first choice given the input interfaces of the ML models, is not the most effective one in all situations.

We study how bank regulation interacts with network topology to influence systemic stability. Employing unique hand‐collected data on the correspondent network for all U.S. banks prior to the Great Depression and a methodology that captures bank credit risk and network position, we demonstrate how the pyramid‐shaped network topology was inherently fragile and systemically risky. We measure its contribution to banking distress, and show that a bank's network position as well as its network neighbors' risk are strong predictors of bank survivorship. Institutional alternatives, such as branch banking, and alternative topologies deliver networks that are more stable than that of 1929.

We present a simple and effective methodology for the generation of lexicons (word lists) that may be used in natural language scoring applications. In particular, in the finance industry, word lists have become ubiquitous for sentiment scoring. These have been derived from dictionaries such as the Harvard Inquirer and require manual curation. Here, we present an automated approach to the curation of lexicons, which makes automatic preparation of any word list immediate. We show that our automated word lists deliver comparable performance to traditional lexicons on machine learning classification tasks. This new approach will enable finance academics and practitioners to create and deploy new word lists in addition to the few traditional ones in a facile manner.

Understanding the predictions made by machine learning (ML) models and their potential biases remains a challenging and labor-intensive task that depends on the application, the dataset, and the specific model. We present Amazon SageMaker Clarify, an explainability feature for Amazon SageMaker that launched in December 2020, providing insights into data and ML models by identifying biases and explaining predictions. It is deeply integrated into Amazon SageMaker, a fully managed service that enables data scientists and developers to build, train, and deploy ML models at any scale. Clarify supports bias detection and feature importance computation across the ML lifecycle, during data preparation, model evaluation, and post-deployment monitoring. We outline the desiderata derived from customer input, the modular architecture, and the methodology for bias and explanation computations. Further, we describe the technical challenges encountered and the tradeoffs we had to make. For illustration, we discuss two customer use cases. We present our deployment results including qualitative customer feedback and a quantitative evaluation. Finally, we summarize lessons learned, and discuss best practices for the successful adoption of fairness and explanation tools in practice.

Credit ratings are traditionally generated using models that use financial statement data and market data, which is tabular (numeric and categorical). Practitioner and academic models do not include text data. Using an automated approach to combine long-form text from SEC filings with the tabular data, we show how multimodal machine learning using stack ensembling and bagging can generate more accurate rating predictions. This paper demonstrates a methodology to use big data to extend tabular data models, which have been used by the ratings industry for decades, to the class of multimodal machine learning models.