[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Graphing quadratic relationship

From   Austin Nichols <[email protected]>
To   [email protected]
Subject   Re: st: Graphing quadratic relationship
Date   Fri, 18 Sep 2009 23:14:01 -0400

Stephanie L Kent:
You might also try -locpr- on SSC.

On Fri, Sep 18, 2009 at 2:34 PM, Ronan Conroy <[email protected]> wrote:
> On 18 MFómh 2009, at 18:35, Stephanie L Kent wrote:
>> Dear users,
>> I would like to graph a relationship between a quadratic independent
>> variable and my dependent variable to see how y varies acccording to x and
>> x^2.  It's a logistic regression so my DV is 0-1 and I need to show how
>> both the 0's and 1's vary according to x and x^2.  Any advice on how to
>> get
>> started is much appreciated!
> I would be wary of quadratic terms, which tend to produce nonsense estimates
> at the extremes of the data, or to extrapolate to nonsense estimates beyond
> the observed range. Have you tried fractional polynomials?
> As to graphing, I have two suggestions:
> 1. jitter your response variable to give a better impression of the density.
> If you have a lot of data, you can end up with two 'stripes' across the
> graph at y=0 and y=1 otherwise
> 2. Add a plot of predicted probabilities as a connected line

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index