****************
****************
** mkspline   **  Using linear splines
****************
****************

  Overview:  The "mkspline" command creates a collection of variables
             that permit a so-called piecewise linear model.  Such a 
             model assumes that the regression is linear between 
             specified points, and that the regression is connected
             at these points. 

             For example, in the framingham data I analyze BMI (body mass 
             index) and wanted to see if a linear relationship is 
             adequate.  I wanted to let the model look something like:


                    |               
                    |               O
                    |            *     *
                    |         *          *
                    |      O                *
                    |     *         
       logit(p)     |    *           
                    |   *   
                    |  *  
                    | * 
                    |*
                    |------|---------|-----------
                           24        28
                              BMI

             where the risk (logit) increases linearly over each range
             but the slope over each range may be different. 


  Usage:     "mkspline  bmiS1 24 bmiS2 28 bmiS3 = bmi"

             "logistic chd bmiS1-3"

             Note that the useage of the "mkspline" command requires that
             you name the variable in each interval, and that you specify
             the values at which you are happy to let the function bend.

             In the example call above "bmiS1" is the linear term before
             "24", bmiS2 is the linear term between "24" and "28", and
             "bmiS3" is the linear term after "28". 

             The coefficients for the logistic regression model are the
             slopes (changes in log odds) over the interval that the 
             variable corresponds to. 

  Summaries: The "mkspline" function creates the variables that you need
             to use this flexible linear model.  The created variables
             can be used for any analysis.

  Options:   (1) "mkspline   bmiS1 24 bmiS2 28 bmiS3 = bmi, marginal"

                  The "marginal" option means that the coefficients 
                  that are fit in the spline model can be interpreted as
                  changes in slopes from the preceeding interval.  This
                  is useful for testing whether the slope is changing
                  from one interval to the next.


****************
****************
** fracpoly   **  Using fractional polynomials
****************
****************

  Overview:  The "fracpoly" command allows use of general polynomial
             regression models (even using fractional powers such 
             as sqrt(X) = X^{1/2}, and negative powers such as
             1/X = X^{-1}).
             This allows a very flexible curve to be used for a 
             continuous covariate.

  Usage:     "fracpoly  logistic chd bmi"

             "fracplot bmi"

             Note that the useage of "fracpoly" comes before a general
             regression command.

  Summaries: The "fracpoly" function actually uses a variable selection 
             method to chose the powers of X that will be used.  This is
             why the fitting procedure can be somewhat slow.  When a model
             is determined the routine will tell you the powers of X that
             are used.

             I don't find the coefficients of these variables interesting,
             but the shape/form can be shown using the "fracplot bmi"
             command.  This includes the regression model with standard error
             bars.


  Options:   (1) "fracplot bmi" -- shows the fitted curve with 95% CIs.