Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Suppose we are looking at an ordinal variable with $4$ categories. So there are three threshold coefficients $b_1, \dots b_3$ and one probit slope $b_4$. What is the interpretation of the following expressions: $$\frac{b_{1}}{b_{4}^{2}}, \frac{b_{2}}{b_{4}^{2}}, \frac{b_{3}}{b_{4}^2}, \frac{1}{b_{4}}$$

assuming a multivariate normal distribution?

share|improve this question
3  
In what sense are you assuming a multivariate response distribution? Typically we think there is a univariate normal response distribution that has been categorized by replacing the value of the latent variable with an ordinal category based on where the value is relative to the thresholds. – gung Jan 4 '13 at 20:16
    
Related: stats.stackexchange.com/questions/46851/…. – whuber Jan 4 '13 at 21:36
    
Why are you looking at these expressions? The threshold divided by the square of the slope? Why would that be interesting? – Peter Flom Jan 4 '13 at 23:23

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.