Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# RE: st: Quantile regression

 From Vasan Kandaswamy To "statalist@hsphsun2.harvard.edu" Subject RE: st: Quantile regression Date Wed, 19 Sep 2012 21:33:33 +0000

```Many thanks Nick.
Now, I have given up on ANOVA since I cannot derive p values for gender seperately, but did a regression.

A quantile regression this way comes up this way
bysort bmi_q sex:sum g0mmol
bysort sex: qreg bmi fast_glucose age pr ( adjusted for age)

I tabulate the output this way
BMI                Q1      Q2        Q3        Q4     Beta (95%CI)            P value
Male              5.3     5.4        5.5        5.6     2.61 (1.46, 3.76)     8.91 x 10^-06
Female         5.4      5.4       5.4         5.7    0.36 (-0.15, 0.86)     0.168

IF you actually look at the mean glucose values in Q1-Q5, there is not much difference, but the regression shows a clear difference with p values of males significant, while females are not.

Could you please explain of my approach is correct.
The basic question I would like to ask is if the fold change from Q1 to Q5 is significant.

Best regards,
Vasan
________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of Nick Cox [njcoxstata@gmail.com]
Sent: Wednesday, September 19, 2012 9:15 PM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: derive exact p values from ANOVA models

1. For every -anova-, there is an equivalent -anova, regress-.

2. 2*normal(-abs(beta/se)) is not "exact" for regression; it is a rule
of thumb even if all assumptions are satisfied, as t statistics are
involved.

3. See Maarten Buis' Tip: full reference and .pdf at

http://www.stata-journal.com/article.html?article=st0137

On Wed, Sep 19, 2012 at 8:03 PM, Vasan Kandaswamy
<vasan.kandaswamy@ki.se> wrote:
> Dear Jay,
>
> Thank you very much. I have now used the two way anova for comparison. Since the groups are equal sizes, regression model is not thought of at the moment.
>
> I have another quick question, is there a way that I could obtain the exact p values from anova.
> I do not want to show p=<0.0001 for all variables, but would like to be more specific.
> While I do regression models, I use beta and SE to derive exact p values this way -
> di (2*normal(-abs(beta/se))). but this is not possible with ANOVA.
>
> Could someone suggest how to get exact p values from ANOVA ?
> Many thanks !
>
> Vasan,
> PhD student,
>
>
> ________________________________________
> From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of JVerkuilen (Gmail) [jvverkuilen@gmail.com]
> Sent: Wednesday, September 19, 2012 1:01 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: ANOVA for quartiles
>
> On Tue, Sep 18, 2012 at 6:08 PM, Vasan Kandaswamy <vasan.kandaswamy@ki.se>wrote:
>
>> Dear Statalisters,
>>
>> I intend to perform an ANOVA to compare the means of a variables across
>> quartiles of BMI ( body mass index) in two genders (male and female ).
>
> Why not a two-way model?
>
>           anova myoutcomevar gender##bmi_quartiles
>
> Or the equivalent sequence of regression models if the groups are not
> of equal size.
>
*
*   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/
```