Skip to main content
. 2014 Feb 10;8:6. doi: 10.3389/fncom.2014.00006

Figure 7.

Figure 7

Fitting mGLM model as opposed to Separate Univariate GLMs. For the example in Data Analysis, the mGLM algorithm achieves a smaller value of the deviance (larger value of the likelihood), but is slower than the algorithm which fits separate univariate GLMs. Comparison of the coefficient estimates would reveal that the latter algorithm underestimates the coefficients corresponding to the simultaneous “11” event.