Discussion:
multinomial output
(too old to reply)
A.A.A
2008-10-02 18:17:33 UTC
Permalink
Hi,

In the spss multinomial logistic regression output i noticed that the
output looks as follows(MY EXPLANATORY VARIABLES ARE BINARY VARIABLES
CODED AS ZERO OR ONE):

Parameter Estimates
V14(a) B

2 Intercept -1.792
[@1=0] -19.627
[@1=1] 0(c)
[@2=0] 2.485
[@2=1] 0(c)
3 Intercept -19.632
[@1=0] 18.785
[@1=1] 0(c)
[@2=0] -18.703
[@2=1] 0(c)
a The reference category is: 1.
b Floating point overflow occurred while computing this statistic. Its
value is therefore set to system missing.
c This parameter is set to zero because it is redundant.


Does this mean the the resulting parameters correspond to the value
zero of the explanatory variable not 1

"Is the value "1" of the explanatory variable set as a reference?Can
this be changed to make the zero the reference?"

Thanks

A.A.A
k***@gmail.com
2020-02-26 15:00:09 UTC
Permalink
Post by A.A.A
Hi,
In the spss multinomial logistic regression output i noticed that the
output looks as follows(MY EXPLANATORY VARIABLES ARE BINARY VARIABLES
Parameter Estimates
V14(a) B
2 Intercept -1.792
3 Intercept -19.632
a The reference category is: 1.
b Floating point overflow occurred while computing this statistic. Its
value is therefore set to system missing.
c This parameter is set to zero because it is redundant.
Does this mean the the resulting parameters correspond to the value
zero of the explanatory variable not 1
"Is the value "1" of the explanatory variable set as a reference?Can
this be changed to make the zero the reference?"
Thanks
A.A.A
Continue reading on narkive:
Loading...