VTechWorks staff will be away for the Thanksgiving holiday beginning at noon on Wednesday, November 27, through Friday, November 29. We will resume normal operations on Monday, December 2. Thank you for your patience.

Browsing by Author "Baker, Matthew"

Now showing 1 - 2 of 2
  • Thumbnail Image
    The Impact of Bankruptcy Exemptions for Retirement Assets
    Baker, Matthew (Virginia Tech, 2013-05-21)
    When filing for personal bankruptcy, an individual can, in almost all cases, claim an exemption for retirement assets.  Using the Survey of Consumer Finances from 2007 and 2010, we test the theory that highly educated or financially sophisticated households allocate more resources to retirement assets under conditions of higher probability of filing for personal bankruptcy.  This hypothesis stems from the concept of asset sheltering, in which an individual will demonstrate a preference for assets that are exempt from a particular risk. To address our hypothesis, we run a Heckman model on the Survey of Consumer Finances data.  Our results provide evidence to match our theory for only highly educated or financially sophisticated individuals, conditional on owning retirement assets.  That is, we observe highly educated and financially sophisticated households allocate more resources to retirement accounts when they are at higher risk for bankruptcy.  Other characteristic groups do not demonstrate a similarly strong relationship between the probability of filing for bankruptcy and the level of retirement assets.
  • Thumbnail Image
    Preparing for the next pandemic via transfer learning from existing diseases with hierarchical multi-modal BERT: a study on COVID-19 outcome prediction
    Agarwal, Khushbu; Choudhury, Sutanay; Tipirneni, Sindhu; Mukherjee, Pritam; Ham, Colby; Tamang, Suzanne; Baker, Matthew; Tang, Siyi; Kocaman, Veysel; Gevaert, Olivier; Rallo, Robert; Reddy, Chandan K. (Nature Portfolio, 2022-06-24)
    Developing prediction models for emerging infectious diseases from relatively small numbers of cases is a critical need for improving pandemic preparedness. Using COVID-19 as an exemplar, we propose a transfer learning methodology for developing predictive models from multi-modal electronic healthcare records by leveraging information from more prevalent diseases with shared clinical characteristics. Our novel hierarchical, multi-modal model (TransMED) integrates baseline risk factors from the natural language processing of clinical notes at admission, time-series measurements of biomarkers obtained from laboratory tests, and discrete diagnostic, procedure and drug codes. We demonstrate the alignment of TransMED's predictions with well-established clinical knowledge about COVID-19 through univariate and multivariate risk factor driven sub-cohort analysis. TransMED's superior performance over state-of-the-art methods shows that leveraging patient data across modalities and transferring prior knowledge from similar disorders is critical for accurate prediction of patient outcomes, and this approach may serve as an important tool in the early response to future pandemics.