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: Predicting sdres in stata


From   Nick Cox <[email protected]>
To   [email protected]
Subject   Re: st: Predicting sdres in stata
Date   Wed, 20 Jul 2011 09:10:08 -0500

No. Perhaps you should study the documentation more, especially that
in [U], on robust variance estimation in Stata. It is not identical to
robust regression. But even with regression, normality of residuals is
usually the least important assumption. As you have bivariate
regressions, plotting scatter and residual plots is just about the
most important checking you can do.

Nick

On Wed, Jul 20, 2011 at 9:01 AM, Lars Folkestad
<[email protected]> wrote:
> Thank you both of you.
> Last question: The robust option, does This render the test of residual normality unnessesery?
>
> Mvh
> Lars Folkestad
>
>
> Den 20/07/2011 kl. 15.34 skrev "Nick Cox" <[email protected]>:
>
>> Note that -predict- without options gives you predicted values, What
>> you call the variable makes no difference to that.
>>
>> Nick
>>
>> On Wed, Jul 20, 2011 at 7:40 AM, Richard Goldstein
>> <[email protected]> wrote:
>>> I think regression is the best way; I am not familiar with how either of
>>> the two concepts are measured; for general guidance on this kind of
>>> adjustment, I suggest the following two articles (which have different
>>> but related points):
>>>
>>> Rosenbaum, PR and Rubin, DB (1984), "Difficulties with regression
>>> analyses of age-adjusted rates," _Biometrics_ 40: 437-443
>>>
>>> Kronmal, RA (1993), "Spurious correlation and the fallacy of the ratio
>>> standard revisited," _Journal of the Royal Statistical Society, Series
>>> A_, 156(3): 379-392; comments and reply in the same journal (1995),
>>> 158(3): 619-625
>>>
>>> Rich
>>>
>>> On 7/20/11 8:28 AM, Lars Folkestad wrote:
>>>> Thank You For the swift answare.
>>>> I was indeed trying to predict the residuals for the regression model.
>>>>
>>>> What i am trying to do is to adjust a Bone Density Value for the
>>>> participants Body surface area. Is there a better way to do this than
>>>> regression?
>>>>
>>>> Will figure wich option fits best.
>>>>
>>>> Lars
>>>>
>>>> Den 20/07/11 14.19 skrev "Richard Goldstein" <[email protected]>
>>>> følgende:
>>>>
>>>>> without knowing what depenVar1 and depenVar2 are, it is not possible to
>>>>> fully answer the question
>>>>>
>>>>> however, note that what you are asking for are the predicted values from
>>>>> the equation and this depends solely on the value of the constant and
>>>>> the value of the coefficient for BSA; apparently, these are "very
>>>>> similar" in the two regressions; do you mean to ask for the predicted
>>>>> values or are you trying to predict some kind of residual? if you want
>>>>> some kind of residual, you will need to add an option; see -h regress
>>>>> postestimation- and click on "predict"
>>>>>
>>>>> Rich
>>>>>
>>>>> On 7/20/11 8:05 AM, Lars Folkestad wrote:
>>>>>> Hi Stata Listers
>>>>>>
>>>>>> This is probably a simple question for you all. I just cannot see my way
>>>>>> through it.
>>>>>>
>>>>>> I am doing liniar regression for different variables as a way to adjust for
>>>>>> Body Surface Area. I do the following
>>>>>>
>>>>>> . regress depenVar1 BSA, vce(robust)
>>>>>> . predict sdres
>>>>>> . qnorm sdres
>>>>>> . swilk sdres
>>>>>> . predict adjdepenVar1
>>>>>> . drop sdres
>>>>>>
>>>>>> . regress depenVar2 BSA, vce(robust)
>>>>>> . predict sdres
>>>>>> . qnorm sdres
>>>>>> . swilk sdres
>>>>>>
>>>>>> The two swilks tests give the exact same p-value and the qnorm graf is
>>>>>> identical.
>>>>>>
>>>>>> I cannot understand how. For your information i am new to stata and
>>>>>> regression and my statistically knowledge is low.
>>>>>>
>>>>>> Why is the two swilks tests and qnorms the same?

*
*   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