STA313f04: Email Questions, with answers

> On the final exam, will there be questions involving R or SAS in any way?

Definitely. But you will not be required to write code during the final
exam.


> While I was printing up the Random Matrices handout, I
> noticed that you skipped page 77 of the text.  Was
> this intentional or an accident?

It looks like a mistake to me. I will scan that page and re-post when I
get home. Thanks for telling me about this.


> It is simple to do a 2nd derivate test for a function
> of 1 variable. Is it simple to do for a function of 3
> variables as well?

In general, no. You check the determinant of a matrix of mixed partial
second derivatives (the Hessian) evaluated at the potential MLE.

HOWEVER, in the case of Question 10, you can do it one variable at a
time. Try maximizing over sigma_xx first, then b, then sigma_uu.

----------------------------------------------------------------------------

Date: Tue, 21 Sep 2004 13:03:29 -0400 (EDT)

> hi sir,
>     how can l eneter the data directly in r without
> using the scan comment?

You could use the c (collect) function illustrated in the first handout,
but that would be very inefficient.

>     if we can't do it then how can l enter my data if
> l save the data in a.txt file?when l save the data in
> txt file ,do l need the fist line i.e Geometric?

To answer your last question first, no.

Suppose you have your data in a file called mydata.txt. There is NOTHING
in this file except numbers. If you are using unix or linux, the data file
should be in the same directory from which you start R. In the MS Windows
version, I don't know for sure, but based on the Mac version of R, I
believe it should be in the R "Working Directory." In the Mac version of
R, there is a Tools menu which contains (among other things) the choices

                        Change Working Directory
                        Show Working Directory
                        Reset Working Directory

Judging from how the R development team usually works, there should also
be a Tools menu in the Windows version, and it should contain the same
choices.

Anyway, once the data are in the right place, at the R prompt type
something like

                x <- scan("mydata.txt")

This should do it.  Best of luck,

Jerry

-------------------------------------------------------------------------------------------