Studying different approaches of estimating linear regression models, I recently came across quantile regression.

In the paper I was reading (http://www.econ.uiuc.edu/~roger/research/rq/rq.pdf) the author uses the following notation:

${\beta} \in \mathbb{R}^{p}$

What does the ${p}$ stand for in this context? I have not come across this notation before.

Thank you.

  • 1
    $\begingroup$ $\beta$ is a $p$-dimensional vector or tuple taking real values. For example if $p=3$, you would be talking about an ordered triple of real numbers such as $(-1,4,-1)$ $\endgroup$ – Henry Feb 19 '18 at 10:32
  • 1
    $\begingroup$ If $p \in \mathbb{N}$ then ${\beta}$ is a $p$-vector. If not, then ${\beta}$ is function from a set $p$ to the set $\mathbb{R}$. $\endgroup$ – Bernard Massé Feb 19 '18 at 10:35
  • $\begingroup$ Thank you for the quick responses. $\endgroup$ – shenflow Feb 19 '18 at 10:41

In quantile regression and more generally in statistics, $p$ often denotes that number of dependent variables (or covariates) in the data. The covariate vector $x_i$ in the paper you read is also $p$-dimensional with component $x_{ij}$ being the value of the $j$th covariate from the $i$th observation, where $i=1,\ldots,n$ and $j=1,\ldots,p$. Here, $n$ often means the number of observations in the data.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.