Dear Marvelous
The following is an extract from How do I interpret odds ratios in logistic regression? (http://www.ats.ucla.edu/stat/mult_pkg/faq/general/odds_ratio.htm.)
......The transformation from probability to odds is a monotonic transformation, meaning the odds increase as the probability increases or vice versa. Probability ranges from 0 and 1. Odds range from 0 and positive infinity
If the probability is 0.001, the odds is 0.001001; if the probability is 0.9999 then the odds is 9999.000... You got an odds value of 243200659, which means that the probability is very high, much more than 0.999999, very near 1; ie your data was like that. It will become higher if you remove the outliers. You got the value because there was very high correlation in your data.
Read the article from the site above. It explains a lot
Narayanan
Nope, not a reasonable estimate at all. Have you checked the integrity of the data? Do you have values that are unreasonable do to data entry or some other reason?
No, it sounds like an inflated value. This can be do to several causes. One could be having two variables that are highly correlated. Another reason could be outliers. Do some exploratory statistics such as box-plots to find if you have outliers. Another reason could be magnitude differences of the variables that you are comparing. How data is coded could cause an inflated odds ratio. You may have some variables that are continuous, but not normally distributed. For example, many biomarkers such as hemoglobin A1C, cholesterol, blood glucose, have outliers and once these are remove, still need to be transformed (usually by natural logarithm) in order to achieve linearity and normality. I suggest you check your independent variables using exploratory statistics. You can use the KS test for a single variable (non-parametric tests) to see if your data is normally distributed or not. Once you transform a variable, you will have to repeat the KS test to see if it is normally distributed. If you cannot achieve normality/linearity for a variable, then you may have to categorize it. Also check all independent variables that are continuous by correlation. Any correlation 0.8 or above may be considered too highly correlated and one should be dropped.
Thank you very much I will attempt to do that and will give you feedback
Dear Marvelous
The following is an extract from How do I interpret odds ratios in logistic regression? (http://www.ats.ucla.edu/stat/mult_pkg/faq/general/odds_ratio.htm.)
......The transformation from probability to odds is a monotonic transformation, meaning the odds increase as the probability increases or vice versa. Probability ranges from 0 and 1. Odds range from 0 and positive infinity
If the probability is 0.001, the odds is 0.001001; if the probability is 0.9999 then the odds is 9999.000... You got an odds value of 243200659, which means that the probability is very high, much more than 0.999999, very near 1; ie your data was like that. It will become higher if you remove the outliers. You got the value because there was very high correlation in your data.
Read the article from the site above. It explains a lot
Narayanan
I think you have received excellent advice. As you can see, it's very important to first check your data to ensure the integrity of the data and assess distributional assumptions. Then you run your models and check for model fit using post-estimation tests.
- Badges
- Science topic
- Similar topics
- References