I am analyzing a data set in which the outcome variable is with count data with large variance, the maximum of which is 84751821 and the minimum is 1. The predictor variables are continuous in nature. When I am trying to do a generalized linear model Poisson's regression, the output i am getting is that there are no valid cases. Statistics cannot be computed. What might be the reason for this output?
You should check your data matrix again; probably there may be any void cases. You may follow this protocol (link attached).
https://statistics.laerd.com/spss-tutorials/poisson-regression-using-spss-statistics.php
It may be helpful. Thanks.
Thanks for your reply. I have gone through the link mentioned by you, but it was not that helpful. I would really like to know if people who have actually computed their data by Poisson regression in SPSS faced a similar problem. And if yes, how can it be resolved?
Hello Dr. Bhattacharjee: I have used the Poisson regression with SPSS. I would like to suggest you first to check your dataset first. The reason is, as suggested by @Sk. Abdul Amin, maybe there is a possibility of some missing values. If any, just check the 'Exclude missing value' option from the option box. This will disobey all those missing values and will generate the matrix based on the non-missing cases only. If there is no missing value, then I guess there might be an ambiguity in selecting your factor and/or the covariates. If you could please specify what exact problem you are facing, like if there is any syntax error or something, that would help us to understand where exactly you are stuck.
~Subhabrata
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS