Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

1. talk about the ar1x process, and your simulation, GLM might not be as good as OLS for estimating parameters, but be careful about inference, why? you use do inference based on …

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

The basic idea is to borrow information across genes. Some assume a distribution for the variances/dispersions while others not.

** Things on this page arefragmentary and immature notes/thoughts of the author. It is not meant to readersbut rather for convenient reference of the author and future improvement. **

1. dirichlet distribution, circle, unit, identical, mixing problem ...

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

QR Decomposition

The QR decomposition uses the Gram-Schmidt process to express a set of vectors (columns of design matrix from statistical view) in a set of orthogonal unit vecotors (which means …

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

The problem is actually to estimate the number of null hypotheses.

• Benjamini

• Nettleton

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

Regression refers to problems where the response (output) variable is continous while classfication refers to problems where the response (output) variable is discrete.

Generally speaking fitting gression to classification problems is …