Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: documentation on iteration for a non linear regression


From   William Buchanan <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: documentation on iteration for a non linear regression
Date   Wed, 13 Jun 2012 07:33:36 -0700

Check out the part of the matrix documentation regarding subscripting individual elements.  In your specific case the matrix is not a scalar, so you would either need to create a series of scalars from individual elements or create macros from those elements.  

help mata

Will provide you with a great deal of information about the mata programming language.  Depending on your specific needs, it may be more beneficial to perform your operations in mata using the matrices and not the individual elements.

- Billy

Sent from my iPhone

On Jun 13, 2012, at 6:40, tashi lama <[email protected]> wrote:

> 
> Thanx a lot. I have none experience working with the matrix in stata. Looks like this is a start. Is there a way to use those matrix contents as a macro or a scalar while writing a do file like we use `r(sum)' after summ varlist command? 
> 
> 
> 
> Thanx tons
> ----------------------------------------
>> Subject: Re: st: documentation on iteration for a non linear regression
>> From: [email protected]
>> Date: Tue, 12 Jun 2012 15:21:02 -0700
>> To: [email protected]
>> 
>> Hi Tashi,
>> 
>> In addition to the requirement that you read the FAQ and try searching for answers in the manuals, your response indicates that you didn't bother to look "in" the stored matrix but instead just looked "at" it. As well as reading the documentation that Nick recommended, I would also suggest reading the documentation regarding how to use matrices in Stata. What you were commenting about is correct, you saw the dimensions of the matrix that you are interested in. Had you entered:
>> 
>> mat li e(b)
>> 
>> You would have displayed the information in the matrix rather than the dimension of the matrix.
>> 
>> - Billy
>> 
>> Sent from my iPhone
>> 
>> On Jun 12, 2012, at 14:40, tashi lama <[email protected]> wrote:
>> 
>>> ereturn list will will provide e(b) under matrices like
>>> 
>>> ereturn list
>>> 
>>> scalars:
>>> 
>>> macros:
>>> 
>>> matrice:
>>> 
>>> e(b): 1*2
>>> 
>>> 
>>> 
>>> that doesn't tell much, does it. Looks like a size of a matrix. I was looking for the the values of parameters b0 and b1 in the table saved as macro or scalar.
>>> 
>>> 
>>> 
>>> Source | SS df MS
>>> -------------+------------------------------ Number of obs = 312
>>> Model | 49079.3588 2 24539.6794 R-squared = 0.9785
>>> Residual | 1077.64118 310 3.47626187 Adj R-squared = 0.9784
>>> -------------+------------------------------ Root MSE = 1.864474
>>> Total | 50157 312 160.759615 Res. dev. = 1272.15
>>> 
>>> ------------------------------------------------------------------------------
>>> hits | Coef. Std. Err. t P>|t| [95% Conf. Interval]
>>> -------------+----------------------------------------------------------------
>>> /b0 | 632.7244 16.97061 37.28 0.000 599.3323 666.1166
>>> /b1 | .3304635 .0074969 44.08 0.000 .3157122 .3452148
>>> ------------------------------------------------------------------------------
>>> 
>>> 
>>> 
>>> 
>>> 
>>> 
>>> 
>>> Thanx
>>> 
>>> 
>>> 
>>> 
>>>> Subject: Re: st: documentation on iteration for a non linear regression
>>>> From: [email protected]
>>>> To: [email protected]
>>>> 
>>>> On Tue, Jun 12, 2012 at 10:09 PM, tashi lama <[email protected]> wrote:
>>>> 
>>>> Is there a documentation in Stata which tells how iteration for a non
>>>> linear regression is done? I am having trouble with initializing the
>>>> parameters. I am hoping I will have a better idea to initialize the
>>>> parameters if I understand the iteration process.
>>>> 
>>>>>>> See the help and the manual entry as usual. The iteration is faster if you have even a rough idea of what the correct values should be.
>>>> 
>>>> Also, when I run a non linear regression model, the values of the
>>>> estimated parameters aren't returned. Isn't it a choke?
>>>> 
>>>>>>> I don't know what "choke" means here, but this is utterly incorrect. Look in e(b).
>>>> *
>>>> * For searches and help try:
>>>> * http://www.stata.com/help.cgi?search
>>>> * http://www.stata.com/support/statalist/faq
>>>> * http://www.ats.ucla.edu/stat/stata/
>>> *
>>> * For searches and help try:
>>> * http://www.stata.com/help.cgi?search
>>> * http://www.stata.com/support/statalist/faq
>>> * http://www.ats.ucla.edu/stat/stata/
>> 
>> *
>> * For searches and help try:
>> * http://www.stata.com/help.cgi?search
>> * http://www.stata.com/support/statalist/faq
>> * http://www.ats.ucla.edu/stat/stata/                         
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index