Publications

Journal Articles

  • J. T. Chi, E. C. Chi, and R. G. Baraniuk, "$k$-POD: A Method for $k$-Means Clustering of Missing Data," The American Statistician, vol. 70, iss. 1, pp. 91-99, 2016.
    [BibTeX]
    @ARTICLE{ChiChiBaraniuk2016,
    author = {Jocelyn T. Chi and Eric C. Chi and Richard G. Baraniuk},
    title = {$k$-POD: A Method for $k$-Means Clustering of Missing Data},
    journal = {The American Statistician},
    year = {2016},
    volume = {70},
    pages = {91-99},
    note = {Available online: [\href{http://www.tandfonline.com/doi/abs/10.1080/00031305.2015.1086685}{e-print},
    \href{http://www.tandfonline.com/doi/pdf/10.1080/00031305.2015.1086685}{pdf}]},
    doi = {10.1080/00031305.2015.1086685},
    issue = {1},
    url = {http://www.tandfonline.com/doi/abs/10.1080/00031305.2015.1086685?journalCode=utas20&}
    }
    [abstract]
    The $k$-means  algorithm  is  often  used  in  clustering  applications  but  its  usage  requires  a complete data matrix.  Missing data, however, is common in many applications.  Mainstream approaches  to  clustering  missing  data  reduce  the  missing  data  problem  to  a  complete  data formulation  through  either  deletion  or  imputation  but  these  solutions  may  incur  significant costs.  Our $k$-POD method presents a simple extension of $k$-means clustering for missing data that works even when the missingness mechanism is unknown, when external information is unavailable, and when there is significant missingness in the data.
    [arXiv] [e-print] [software]
  • J. T. Chi and M. S. Handcock, "Identifying Sources of Healthcare Underutilization Among California's Immigrants," Journal of Racial and Ethnic Health Disparities, vol. 1, iss. 3, pp. 207-218, 2014.
    [BibTeX]
    @ARTICLE{ChiHandcock2014, 
    author = {Jocelyn T. Chi and Mark S. Handcock},
    title = {Identifying Sources of Healthcare Underutilization Among California's Immigrants},
    journal = {Journal of Racial and Ethnic Health Disparities},
    year = {2014},
    volume = {1},
    issue = {3},
    pages = {207-218},
    url = {http://link.springer.com/article/10.1007/s40615-014-0028-0},
    doi = {10.1007/s40615-014-0028-0}
    }
    [abstract]
    Many studies show that immigrants face significant barriers in accessing health care. These barriers may be particularly pronounced for newer immigrants, who may face additional obstacles in navigating the health care system.  Understanding the sources of health care disparities between recent and non-recent immigrants may allow for better design of policies and interventions to address the vulnerabilities unique to different subgroups of immigrants defined by their length of residency. This study employs descriptive analyses and multivariate logistic regression to estimate the likelihood of accessing and utilizing health care services based on immigration-related factors after controlling for predisposing, enabling, and health care need factors. We also employ a regression-based decomposition method to determine whether health care differences between recent and non-recent immigrants are statistically significant and to identify the primary drivers of healthcare differences between recent and non-recent immigrants. The findings support the hypothesis that significant disparities in health care access and utilization exist between recent and non-recent immigrants. We found that health care access and utilization differences between recent and non-recent immigrants were driven primarily by enabling resources, including limited English proficiency (LEP), insurance status, public assistance usage, and poverty level. These results indicate that not only are newer immigrants more likely to underutilize health care, but also that their underutilization is driven primarily by their lack of insurance, lack of adequate financial resources, and inability to navigate the health care system due to LEP. The results further indicate that immigrants with prolonged LEP may be less likely to have a usual source of care and more likely to report delays in obtaining medical treatments, than even recent immigrants with LEP.
    [e-print]

Unrefereed Work

  • J. T. Chi and E. C. Chi, “Getting to the Bottom of Matrix Completion and Nonnegative Least Squares with the MM Algorithm,” StatisticsViews, March 2014.
    [BibTeX]
    @MISC{ChiChiMMAlgorithm, 
    author = {Jocelyn T. Chi and Eric C. Chi},
    title = {Getting to the Bottom of Matrix Completion and Nonnegative Least Squares with the MM Algorithm},
    howpublished = {StatisticsViews.com},
    month = {March},
    year = {2014},
    url = {http://www.statisticsviews.com/details/feature/6035321/Getting-to-the-Bottom-of-Matrix-Completion-and-Nonnegative-Least-Squares-with-th.html}
    }
    [e-print] [software]
  • J. T. Chi and E. C. Chi, “Getting to the Bottom of Regression with Gradient Descent,” StatisticsViews, January 2014.
    [BibTeX]
    @MISC{ChiChiGradientDescent, 
    author = {Jocelyn T. Chi and Eric C. Chi},
    title = {Getting to the Bottom of Regression with Gradient Descent},
    howpublished = {StatisticsViews.com},
    month = {January},
    year = {2014},
    url = {http://www.statisticsviews.com/details/feature/5722691/Getting-to-the-Bottom-of-Regression-with-Gradient-Descent.html}
    }
    [e-print] [software]