This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Statistics This article is within the scope of WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.StatisticsWikipedia:WikiProject StatisticsTemplate:WikiProject StatisticsStatistics ??? This article has not yet received a rating on the importance scale.

There is a move discussion in progress on Talk:Student's t-distribution which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 04:01, 24 August 2021 (UTC)[reply]

The "generalized" version of the distribution is actually just a re-parameterization, not an extension of the distribution family defined by the "standard" formulation. The "generalized" form is based on the Inverse Matrix Gamma distribution, which is also just a re-parameterization of the inverse Wishart, just like the Matrix Gamma is another formulation of the Wishart distribution. I suggest we should rename "Generalised" -> "Re-parameterized", and also point out that it is over-parameterised. ArneLeijon (talk) 13:06, 1 July 2024 (UTC)[reply]

Dear previous contributors, Please take a look at my latest edits. I have added conditions for well-defined Mean and Covariance to exist, to be consistent with the (vector-)multivariate t-distribution. I am sure the previous version had an incorrect condition for the Mean in the box for the "standard" Matrix t version.

Gupta & Nagar (2000, Theorem 4.3.1) stated the mean as E[T]= M without any restriction on the degree of freedom parameter, and Iranmanesh et al (2010) states the same with ref to Gupta. However, I still think there are good reasons to keep the statement for the Matrix t consistent with the corresponding condition for the vector-multivariate t-distribution: Since the Matrix t distribution is equivalent to the vector-multivariate version if the matrix has only one row, or only one column, the requirements on the degree of freedom parameter must be the same, and I am sure the article on the (vector-)multivariate t distribution is correct on this point. ArneLeijon (talk) 20:29, 1 July 2024 (UTC)[reply]

Something is off in the PDF because the dimensions of the matrices do not match. My best guess is that Σ and Ω are swapped in the determinant but I don't know if this should be fixed by either swapping the symbols or by making Σ a (n x n) matrix and Ω a (p x p) matrix. DenverCoder91 (talk) 07:58, 28 October 2025 (UTC)[reply]