sas >> logistic regression for continuous dependent variable

by kea 2003 » Wed, 19 Jan 2011 15:29:34 GMT

Dear All,

Is it possible to run logistic regression on a dependent variable
carrying continuous values?

If so,
- are there any theoretical violations?
- do the model estimations become biased?
- any convergence problems with the iterative logistic algorithm?

Many thanks in advance for any insight that you may be able to
provide.

Kind regards,
Kea

sas >> logistic regression for continuous dependent variable

by rjs » Wed, 19 Jan 2011 19:33:31 GMT


Hi Kea

I used to hate this kind of reply, but I have to say that based on
this post and your previous one, I get the feeling that you're not
that familiar with regression or its valid usage. I think that this
isn't particularly a software problem. Rather than go directly to the
SAS capabilities maybe the safest thing for you to do would be take a
big step back and think about the theory and intentions behind what
you're trying to achieve. Without that you won't be able to interpret
your results so there won't really be any point in having them. Maybe
read a couple of chapters about different types of regression (try
William Greene or Peter Kennedy). If possible get some advice from an
experienced statistician too.

Regards

Robert

Similar Threads

1. logistic regression for continuous dependent variable

Dear All,

Is it possible to run logistic regression on a dependent variable
carrying continuous values?

If so,
- are there any theoretical violations?
- do the model estimations become biased?
- any convergence problems with the iterative logistic algorithm?

Many thanks in advance for any insight that you may be able to
provide.

Kind regards,
Kea

2. LOGISTIC REGRESSION OF NOMINAL DEPENDENT (Y) VARIABLE AND ORDINAL PREDICTOR (X) VARIABLE

3. Logistic Regression and Unequal Distribution of Dependent Variable

Hello, I'm preparing to run a logit model predicting the odds of NOT
testing for an STD.

As you can see from the table below, 2934 (about 86%) of respondents have my outcome of interest (i.e., have not tested for an STD).

I realized that because of this unequal unequal distribution of the dependent variable, all crosstabulations have higher proportions within the untested category of those who have not been tested, regardless of the distribution of the other variable.

I have a feeling that these could bias my estimates in a way - since the not-tested category seemed over-estimated. For example, given the unequal groupings, I think I am only restricted to modeling failure to test (the zero outcome), as modeling for ever tested (1) could lead to unstable estimates.

So my question is it worth producing any crosstabs showing the distribution of socio-demographic variables within my outcome of interest?

What possible impact will this have on my logistic model, and what can I do about it?  Thanks - Yawo

===================>
Table 1:

RECODE of |
V827      |
(Last     |
test was  |
on your   |
own,      |
offered   |  RECODE of V501 (Current
or        |      marital status)
required) |     0      1      2  Total
----------+---------------------------
 Not Test | 99.37   81.1  99.08  88.75
          |   514   1563    857   2934
          |
 Asked fo | .2992  1.015  .2525  .6992
          |     2     18      2     22
          |
  Offered | .2523  17.63  .1184  10.24
          |     3    427      1    431
          |
 Test Req | .0816   .253  .5512  .3114
          |     1      5      2      8
          |
    Total |   100    100    100    100
          |   520   2013    862   3395
--------------------------------------
  Key:  column percentages
        number of observations
----------------------------------





Table 2:

RECODE of |
V827      |
(Last     |
test was  |
on your   |
own,      |
offered   |  RECODE of V106 (Highest
or        |     educational level)
required) |     0      1      2  Total
----------+---------------------------
 Not Test | 83.34  96.84   89.9  88.75
          |   724    273   1937   2934
          |
 Asked fo | .2094  1.662   .777  .6992
          |     2      4     16     22
          |
  Offered | 16.37  1.497  8.887  10.24
          |   209      3    219    431
          |
 Test Req | .0785      0  .4358  .3114
          |     1      0      7      8
          |
    Total |   100    100    100    100
          |   936    280   2179   3395
--------------------------------------
  Key:  column percentages
        number of observations

4. Dependent variable in logistic regression

5. Dependent variable is ratio of continuous values

Hi.

I have a situation where the target variable I want to model takes on
values between 0 and 1, endpoints inclusive, because it is a ratio of two
continuous quantities (e.g., actual sales / sales target)...

Is there a "preferred" method for modeling ratios of continuous quantities,
and if so, is it available in SAS?


Someone suggested the use of the PROC LOGISTIC events/trials syntax, but
the SAS documentation for that really seems to stress binary outcome
trials...

Is it legitimate to use PROC LOGISTIC events/trials for continuous
numerator and denominator?


An econometrics textbook ("Econometric Analysis" by W.H. Greene, 5th
edition, Prentice-Hall) suggests Minimum Chi-Squared Estimation (or MCSE --
don't tell Microsoft!) for proportions; it looks like that discussion was
motivated by proportions of binary outcomes, but I think the equations
still work in the case of continuous numerator and denominator.  However, a
search of support.sas.com didn't turn up any procedures that support the
MCSE methodology.  One weakness of MCSE is that the estimation only works
when the proportion does not take on the extreme values of 0 or 1.  In such
cases, it seems that the suggested work-around is to add or subtract a
small constant...

Is there a SAS procedure that does minimum chi-squared estimation for
proportions?


I could also transform the ratio from the unit interval to the entire line
with the logit transform, log(y/(1-y)), and then perform a standard
regression.  Again, I'd have to smudge the extreme values before performing
the transformation...

Is it legitimate to logit transform the ratio after slightly modifying the
extreme values, and then do OLS?


Thanks, as ever, for your indulgence...

--  TMK  --

6. Logistic Regression and Unequal Distribution of Dependent

7. Stat Question--Bucket a continuous variable--Is it still a continuous variable

8. Interaction between categorical and continuous variables in regression model