st: RE: Multiple linear regression the right approach?

["William Buchanan" <william@williambuchanan.net>]

Hi Simon,

You could probably get a much better response if you provided some basic
information about your data to the listserv (e.g., descriptive statistics,
what the data are, etc...).  If one of your independent variables is
categorical it wouldn't have a scale that would be interpretable (e.g., if
you coded Black = 3; White = 2; Asian = 1 it doesn't mean that Whites or
Asians have less of a racial property and the numbers signify nothing); I
assume that you meant that the variable is ordinal in nature (e.g., the
numbers convey magnitude but are not necessarily proportional).  

Which variable do you assume is not linearly related to your dependent
variable?  Are your dependent variable and independent variables measured on
similar scales (in terms of orders of magnitude)? What does the non-linear
relationship appear to be (e.g., quadratic, cubic, quartic, something else,
etc...)? 

It is difficult to provide any useful feedback without knowing more of these
details and any feedback at this point could lead you to the same place.  

HTH,
Billy

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Simon Hauburger
Sent: Monday, June 24, 2013 11:47 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: Multiple linear regression the right approach?

Dear potential helpers,

I have a problem figuring out the right regression for my model:

- It has a interval dependent variable (costs in $)  that looks normally
distributed, but according to shapiro-wilk test isn't
- a number of independent variables which are categorial (scale from
1-6) and interval (assets in $)

My first guess was to use a multiple linear regression, but not all of the
independent variables are linearly related to the dependent variable (tested
with cprplot lowess), even after having tried the common transformation
techniques (log, square...)

Any reommendations for my next steps? Keep trying to transform the variables
and use the multiple linear regression or try an alternative method? If so,
which method could it be? Logistic regression?
(Transformation of the dependent variable to a binary variable is
possible)

I am really confused, statistics will never become my best friend....

Thank you for your help!

Best
*
*   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/