Re: st: test for trend of an ordinal predictor

[Richard Goldstein <richgold@ix.netcom.com>]

first, please respond via statalist (see the faq)

second, -cascade- is designed to set up "special" dummy variables for
ordinal variables; these can then be used in models including regression

I am still not sure exactly what you mean by "trend" so suggest you
either (1) review the stb article or (2) specify, possibly with an
example using, e.g., the auto data, more clearly exactly what your null
hypothesis is

Rich

On 4/26/10 9:02 AM, Georgia wrote:
> I am really sorry. I probably didn't set the question clearly.
> What I want to do is to test for trend across categories of an ordinal
> predictor in a linear regression model. How can I do that?
> 
> On 26/04/2010 13:44, Richard Goldstein wrote:
>> it is not exactly clear what you want here, but you may find the
>> -cascade- program helpful; you can find this using -findit cascade-; the
>> original stb article for this can be found at
>> www.stata.com/products/stb/journals/stb6.html
>>
>> Rich
>>
>> On 4/26/10 8:31 AM, Georgia wrote:
>>   
>>> Hi there,
>>>
>>> I am fitting a regression model with a normally distributed outcome (y)
>>> and an ordinal variable in 3 levels (x) as predictor. Obviously, when I
>>> use the command
>>>
>>> xi: regress y i.x
>>>
>>> I get separate coefficients for the second and the third level of the
>>> predictor while the first level is treated as reference. However, I
>>> would be very interested in testing the trend of the ordinal variable.
>>> In such a case, do I simply treat my ordinal var as a continuous one by
>>> typing
>>>
>>> regress y x
>>>
>>> or is there  a better way to test the trend of x? I would appreciate any
>>> help on this.
>>>
>>> Georgia
>>>      
> 
> 
> 
*
*   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/