Skip to main content
. Author manuscript; available in PMC: 2014 Mar 1.
Published in final edited form as: Biometrics. 2013 Jan 22;69(1):197–205. doi: 10.1111/j.1541-0420.2012.01812.x

Figure 2.

Figure 2

Linear regression models for mean log Epo dose as a function of lagged Epo dose and lagged hematocrit. The potential predictors are expanded to include quadratic terms for lagged Epo and hematocrit, as well as pairwise interactions between lagged variables. Shown is the trace of estimated standardized coeffiients subjected to the LASSO penalty as a function of the L1 norm of the coefficient vector.