search for


Identifying factors associated with university hospitals’ profitability based on fused lasso regression
Journal of the Korean Data & Information Science Society 2018;29:83-96
Published online January 31, 2018
© 2018 Korean Data and Information Science Society.

Ji Hoon Lee1 · Jincheol Park2 · Donghyeon Yu3

12Department of Statistics, Keimyung University
3Department of Statistics, Inha University
Correspondence to: Assistant Professor, Department of Statistics, Inha University, Incheon 22212, Korea. E-mail:
Received September 11, 2017; Revised January 11, 2018; Accepted January 17, 2018.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
This study considers the fused lasso regression model to identify the factors associated with the profitability of 21 university hospitals based on their management information from 2013 to 2015. The fused lasso regression has a grouping property that the coefficients of positively and highly correlated variables are estimated identically. With this property, the fused lasso regression identified several groups of variables and resolved the multicolinearity problem that occurred when we applied the multiple linear regression model. Moreover, we additionally consider three parameters of intercepts in the fused lasso regression model to take into account the changes in the means of yearly-observed variables for three years. From this study, we have identified the factors that negatively impact on medical income operating profit margin ratio: the number of patients treated per nurse, one-day average medical expenses per outpatient, length of stay, employment cost rate, material cost rate, and management expense rate.
Keywords : Fused lasso, lasso, multicolinearity, regression, university hospitals’ profitability.