Since you say it is panel data, I assume that you have repeated binary responses over time from each customer.  Observations within one customer are likely correlated and the correlation should be taken into account in the analysis.  Probably the most common approach for this is a Generalized Estimating Equations (GEE) logistic model which can be done using PROC GEE or PROC GENMOD.  Use the REPEATED statement in either of these, specifying a variable in the SUBJECT= option that has a unique value for each customer.  The data set should have multiple observations per customer.  Specify the DIST=BINOMIAL option in the MODEL statement to fit a GEE logistic model.  Another approach is a conditional logistic model which can be fit in PROC LOGISTIC using your customer variable in the STRATA statement. Either approach allows unequal numbers of responses per customer. Some discussion of these models can be found in: "Categorical Data Analysis Using SAS, Third Edition" (Stokes, M. et. al., SAS Institute, 2012) "Logistic Regression Using SAS: Theory and Application, Second Edition," (Allison, P., SAS Institute, 2012) "Fixed Effects Regression Methods for Longitudinal Data Using SAS"  (Allison, P., SAS Institute, 2005)   ... View more

For a four-level response, only three independent response functions (logits) can be simultaneously modeled - each using a ratio of the probabilities of one response level over the reference level. As a result, you get LS-mean estimates and pairwise comparisons among the predictor levels for each of those three response functions. The easiest way to get the estimates and comparisons for a logit focused on the reference response level is to change the reference level of the response. To do that, use the REF= option following the response variable name in the MODEL statement and specify (in quotes) a value of the response other than the reference level used in your initial analysis.   For example, the following uses the data in the example titled "Nominal Response Data: Generalized Logits Model" in the LOGISTIC documentation. The response, Style, has three levels - self, team, and class. The first step below fits the model to logits for styles self and team using class as the reference in both logits. The probability estimates for each school on the self and team styles are provided by the LSMEANS / ILINK statement and appear in the Mean column in the table of Least Squares Means. The estimates for the class style are provided by the second step which changes the reference level of the response to self instead of class. This code uses PROC LOGISTIC, but the same should work with SURVEYLOGISTIC.   proc logistic data=school;    freq Count;    class School/param=glm;    model Style(order=data)=School / link=glogit;    lsmeans school/diff ilink; run; proc logistic data=school;    freq Count;    class School/param=glm;    model Style(order=data ref="self")=School / link=glogit;    lsmeans school/diff ilink; run;   ... View more

As stated by Clausen (1998) when discussing the interpretation of distances between row and column points, "Usually, however, the points i and j will be close to each other when f(ij)>e(ij), and the distance will be great when f(ij)<e(ij), where f(ij) is the observed and e(ij) is the expected frequency... ."  Intuitively, observed counts larger than expected (see the EXPECTED and CELLCHI2 options in PROC CORRESP) indicate some association and this tends to be depicted visually in the plot by closeness.  But no, the distances are not chi-square distances as the are between two row points or two column points.    Claussen, S-E (1998), Applied Correspondence Analysis: An Introduction, Sage University Papers Series on Qualitative Applications in the Social Sciences, 07-121. Thousand Oaks, CA: Sage. ... View more

You can fit a beta model using PROC GLIMMIX or PROC FMM.  See the DIST=BETA option in the MODEL statement. See this example of using the beta distribution in GLIMMIX to model a continuous proportion response. ... View more

PROC LOGISTIC gives a count of uninformative strata in the Model Information table.  By default, the Strata Summary table shows the various strata profiles.  Uninformative strata have a zero count for one of the two response levels in this table.  The INFO option in the STRATA statement provides a list of the strata if desired. ... View more

The "L'Beta Estimate" and "Standard error" values, and the "L'Beta Confidence limits" in the ESTIMATE statement results are on the logit scale (log[p/(1-p)]) when fitting a logistic model. For an ESTIMATE statement that defines a difference in two logits, using the EXP option provides the odds ratio estimate and its confidence limits. Equivalently, you can use an LSMEANS statement along with the DIFF, ODDSRATIO, and CL options. For example, using the GEE example in the Getting Started section of the GENMOD documentation, adding these statements provides the odds ratio estimate and confidence interval for comparing the two cities:   estimate 'city' city 1 -1 / exp; lsmeans city / diff oddsratio cl;   However, a standard error on the odds ratio scale is not available.   ... View more

Effects parameterization (or coding) compares each level to the average of all of the levels.  Reference parameterization, which you used, compares each level to the reference level.  To use effects coding, specify PARAM=EFFECT. ... View more