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: Checking to see if the association between two variables is linear or otherwise
From
William Buchanan <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: Checking to see if the association between two variables is linear or otherwise
Date
Sat, 13 Oct 2012 07:32:55 -0700
I'm not sure which suggesting you're referencing. All I did was mention that there was an example of a similar idea in one of the Stata Press books and made no qualitative assumptions regarding the merits and/or shortcomings of the methodology.
-Billy
Sent from my iPhone
On Oct 13, 2012, at 0:25, "Justina Fischer" <[email protected]> wrote:
> your suggestion does not resolve the problem that years sof education measure degree very inacurately.
>
> In A US context, 12 and 16 years of education most likely relate to high school degree and bachelor's degree attained four years later.
>
> That Mitchel find splines at these nots suggests that using educational-level-dummies is a very good idea indeed.
>
> Justina
>
> -------- Original-Nachricht --------
>> Datum: Fri, 12 Oct 2012 20:25:54 -0700
>> Von: William Buchanan <[email protected]>
>> An: "[email protected]" <[email protected]>
>> Betreff: Re: st: Checking to see if the association between two variables is linear or otherwise
>
>> It's one of the examples in Michael Mitchell's book on data visualization.
>> He used knots at 12 and 16 years of education to illustrate how you could
>> use splines and the different commands to show the different relationships
>> between years of education and earnings.
>>
>> - Billy
>>
>> Sent from my iPhone
>>
>> On Oct 12, 2012, at 20:04, "JVerkuilen (Gmail)" <[email protected]>
>> wrote:
>>
>>> On Fri, Oct 12, 2012 at 10:46 PM, David Kantor <[email protected]> wrote:
>>>
>>>> I agree, noting that highest grade attained is better than years in
>>>> education, but it is still not appropriate, as it is not an interval
>> scale.
>>>> The "same" increments between values do not have the same significance.
>>>> For example, the significance of the increment from grade 11 to 12 is
>> not
>>>> the same as from 10 to 11.
>>>
>>> Interesting point. I bet this could probably be ameliorated using
>>> regression splines with knots at key grade milestones.
>>>
>>> Jay
>>> --
>>> JVVerkuilen, PhD
>>> [email protected]
>>>
>>> "Out beyond ideas of wrong-doing and right-doing there is a field.
>>> I'll meet you there. When the soul lies down in that grass the world
>>> is too full to talk about." ---Rumi
>>> *
>>> * 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/