Dirichlet component regression and its applications to psychiatric data

  title={Dirichlet component regression and its applications to psychiatric data},
  author={Ralitza Gueorguieva and Robert Rosenheck and Daniel Zelterman},
  journal={Computational statistics & data analysis},
  volume={52 12},
We describe a Dirichlet multivariable regression method useful for modeling data representing components as a percentage of a total. This model is motivated by the unmet need in psychiatry and other areas to simultaneously assess the effects of covariates on the relative contributions of different components of a measure. The model is illustrated using the Positive and Negative Syndrome Scale (PANSS) for assessment of schizophrenia symptoms which, like many other metrics in psychiatry, is… CONTINUE READING


Publications citing this paper.
Showing 1-9 of 9 extracted citations

Profile Monitoring for Compositional Data

Rubén Darío Guevara-González, José Alberto Vargas-Navas, Dorian Luis Linero-Segrera
View 6 Excerpts
Highly Influenced

Functional traits explain variation in plant life history strategies.

Proceedings of the National Academy of Sciences of the United States of America • 2014
View 1 Excerpt

Similar Papers

Loading similar papers…