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: theory reg vs. qreg


From   Nick Cox <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: theory reg vs. qreg
Date   Mon, 29 Apr 2013 12:38:30 +0100

Sorry, nothing here is clear to me.

If you have a point of principle, you need to define your terms.

If you have results from Stata that puzzle you, you can give us an
example, preferably using a dataset we can all access.

Nick
[email protected]


On 29 April 2013 12:30, Kasal Roman <[email protected]> wrote:
> but do you agree with:
>
> if "reg" gives me the general results:
> A) Y1=b1
> B) Y2=a2*x+b2
>
> on the same data
> then must be:
>
> Y2=mean(x)*x+b2=b1=Y1 ?
>
> Is the assumption clear now?
>
> thx
>
> On Mon, Apr 29, 2013 at 1:21 PM, Nick Cox <[email protected]> wrote:
>> I see no evidence here for that extraordinary claim.
>>
>> Sorry, I am still lost on what you want.
>> Nick
>> [email protected]
>>
>>
>> On 29 April 2013 12:17, Kasal Roman <[email protected]> wrote:
>>> of course I don't mean this, I know a2!=mean(x) but IF you put
>>> a2=mean(x) you get the first method result with the "reg" command :)
>>>
>>> On Mon, Apr 29, 2013 at 12:39 PM, Nick Cox <[email protected]> wrote:
>>>>
>>>> What you want is not at all clear. In fact, you seem confused about some basics.
>>>>
>>>>  For example, if you fit a regression
>>>>
>>>> Y2=a2*x+b2
>>>>
>>>> it's not true that a2 = mean(x).
>>>>
>>>> What do you mean by "deactivate" a coefficient?
>>>>
>>>> -qreg y-
>>>>
>>>> works; it's just a round-about way of calculating the median of -y-,
>>>> which you can do directly with -summarize-.
>>>>
>>>> Nick
>>>> [email protected]
>>>>
>>>>
>>>> On 29 April 2013 11:23, Kasal Roman <[email protected]> wrote:
>>>> > Please,
>>>> >
>>>> >
>>>> > the "reg" command gives me these results:
>>>> >
>>>> > A) Y1=b1
>>>> >
>>>> > B) Y2=a2*x+b2
>>>> >
>>>> > then
>>>> >
>>>> > Y2=mean(x)*x+b2=b1=Y1
>>>> >
>>>> >
>>>> > How is it with "qreg"? How could I deactivate the "a2" coefficient
>>>> > without a need to estimate two functions with the "qreg" command? How
>>>> > to get the result of "b1" just with the B) method through the "qreg"
>>>> > command? I know neither mean(x) nor median(x) fit.
>>>> *
>>>> *   For searches and help try:
>>>> *   http://www.stata.com/help.cgi?search
>>>> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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