fracpoly.hlp

是一个经济学管理应用软件 很难找的 但是经济学学生又必须用到

{smcl}
{* 28feb2005}{...}
{cmd:help fracpoly}, {cmd:help fracgen}{right:dialogs:  {bf:{dialog fracpoly}}  {bf:{dialog fracgen}}{space 6}}
{right:also see:  {help fracpoly postestimation}}
{hline}

{title:Title}

{p2colset 5 21 23 2}{...}
{p2col :{hi:[R] fracpoly} {hline 2}}Fractional polynomial regression{p_end}
{p2colreset}{...}


{title:Syntax}

{phang}
Fractional polynomial regression

{p 8 17 2}
{cmd:fracpoly}
{it:{help fracpoly##regression_cmd:regression_cmd}}
{it:yvar}
{it:xvar1}
[{it:#} [{it:#}{it:...}]]
[{it:xvar2} [{it:#} [{it:#}{it:...}]]] [{it:...}]
[{it:xvarlist}]
{ifin}
{weight}
[{cmd:,}
{it:{help fracpoly##fracpoly_options:fracpoly_options}}]

{phang}
Display table showing the best FP model for each degree

{p 8 17 2}{cmd:fracpoly} [{cmd:,} {opt com:pare}]

{phang}
Create variables containing fractional polynomial powers

{p 8 17 2}
{cmd:fracgen}
{it:varname}
{it:#} [{it:#} {it:...}]
{ifin}
[{cmd:,}
{it:{help fracpoly##fracgen_options:fracgen_options}}]

{pstd}
{marker regression_cmd}where {it:regression_cmd} may be {helpb clogit},
{helpb glm}, {helpb logistic}, {helpb logit}, {helpb poisson}, {helpb probit},
{helpb regress}, {helpb stcox}, or {helpb streg}.

{synoptset 26 tabbed}{...}
{marker fracpoly_options}{...}
{synopthdr :fracpoly_options}
{synoptline}
{syntab :Model}
{synopt :{opt deg:ree(#)}}degree of fractional polynomial to fit; default is
{cmd:degree(2)}{p_end}
{synopt :{it:regression_cmd_options}}options appropriate to the regression
command in use{p_end}

{syntab :Model 2}
{synopt :{opt nosca:ling}}suppress scaling of first independent variable{p_end}
{synopt :{opt nocon:stant}}suppress constant term{p_end}
{synopt :{opth po:wers(numlist)}}list of fractional polynomial powers from
which models are chosen{p_end}
{synopt :{opt ad:just(adj_list)}}specification of adjustment for the
independent variables{p_end}

{syntab :Reporting}
{synopt :{opt log}}display iteration log{p_end}
{synopt :{opt comp:are}}compare models by degree{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}
All weight types supported by {it:regression_cmd} are allowed; see 
{help weight}.{p_end}
{p 4 6 2}
See {help fracpoly postestimation}
for features available after estimation.{p_end}
{p 4 6 2}
where {it:adj_list} is a comma-separated list with elements
{it:varlist}{cmd::}{c -(}{opt mean}|{it:#}|{opt no}},
except that the first element may optionally be of the form
{c -(}{opt mean}|{it:#}|{opt no}} to specify the default for all variables.
{p_end}

{synoptset 26 tabbed}{...}
{marker fracgen_options}{...}
{synopthdr :fracgen_options}
{synoptline}
{syntab:Main}
{synopt :{cmdab:a:djust(no}|{cmd:mean}|{it:#}{cmd:)}}specify whether {it:varname} is to be adjusted; default is {cmd:adjust(no)}{p_end}
{synopt :{opt nosca:ling}}suppress scaling of {it:varname}{p_end}
{synopt :{opt replace}}replace variables if they exist{p_end}
{synoptline}
{p2colreset}{...}


{title:Description}

{pstd}
{opt fracpoly} fits fractional polynomials in {it:xvar1} to
{it:yvar}.  After execution, {opt fracpoly} leaves variables in the dataset
named {bf:I}{it:xvar}{bf:__1}, {bf:I}{it:xvar}{bf:__2}, ..., where {it:xvar}
represents the first four letters of the name of {it:xvar1}.  The new
variables contain the best-fitting fractional polynomial powers of {it:xvar1}.

{pstd}
Covariates other than {it:xvar1}, which are optional, are specified in
{it:xvar2},... and {it:xvarlist}.  They may be modeled linearly and
with specified fractional polynomial transformations.  Fractional polynomial
powers are specified by typing numbers following the variable's name.  A
variable name typed without numbers following it is entered linearly.

{pstd}
{opt fracgen} creates new variables named {it:varname}{bf:_1},
{it:varname}{bf:_2},... containing fractional polynomial powers of
{it:varname} using the powers {cmd:(}{it:#}[{it:#} {it:...}]{cmd:)}
specified.

{pstd}
See {help fracpoly postestimation} for information on {opt fracplot} and
{opt fracpred}.

{pstd}
See {helpb mfp} for multivariable fractional polynomial model fitting.


{title:Options for fracpoly}

{dlgtab:Model}

{phang}
{opt degree(#)} determines the degree of fractional
polynomial to be fitted.  The default is {cmd:degree(2)}, i.e., a model with
two power terms.

{phang}
{it:regression_cmd_options} are options appropriate to the regression
command in use.  For example, for {opt stcox}, {it:regression_cmd_options} may
include {opt efron} or some other alternate method for handling tied failures.

{dlgtab:Model 2}

{phang}
{opt noscaling} suppresses scaling of {it:xvar1} and its powers.

{phang}
{opt noconstant} suppresses the regression constant if this is
permitted by {it:regression_cmd}.

{phang}
{opth powers(numlist)} is the set of fractional
polynomial powers from which models are to be chosen.  The default is
{cmd:powers(-2, -1, -.5, 0, .5, 1, 2, 3)} (0 means log).

{phang}
{opt adjust(adj_list)} defines the adjustment for the
covariates {it:xvar1}, {it:xvar2}, ..., {it:xvarlist}.  The default is
{cmd:adjust(mean)}. 
A typical item in {it:adj_list} is
{it:varlist}{cmd::}{c -(}{cmd:mean}|{it:#}|{cmd:no}}.
Items are separated by commas.   The first
item is special in that {it:varlist}{cmd::} is optional, and if omitted, the
default is (re)set to the specified value ({opt mean} or {it:#} or {opt no}).
For example, {cmd:adjust(no, age:mean)} sets the default to {opt no} and sets
the adjustment for {opt age} to {opt mean}.

{dlgtab:Reporting}

{phang}
{cmd:log} displays deviances and (for {helpb regress}) residual standard
deviations for each fractional polynomial model fitted.

{phang}
{cmd:compare} reports significance tests between the best fractional
polynomial models of increasing degree.


{title:Options for fracgen}

{dlgtab:Main}

{phang}
{cmd:adjust(no}|{cmd:mean}|{it:#}{cmd:)}
specifies whether {it:varname} is to be adjusted; the default is
{cmd:adjust(no)}.

{phang}
{opt noscaling} suppresses scaling of {it:varname}.

{phang}
{opt replace} specifies that any variables named {it:varname}{bf:_1},
named {it:varname}{bf:_2}, ... may be replaced.


{title:Examples}

{phang}{cmd:. fracpoly regress sqrtigg age, degree(4) compare}{p_end}

{phang}{cmd:. fracpoly regress sqrtigg age 1 2 3 4}{p_end}
{phang}{cmd:. fracpoly regress sqrtigg age -2 2}{p_end}

{phang}{cmd:. fracgen mpg 0 0}{p_end}
{phang}{cmd:. fracgen mpg -2 -1 if foreign==1, noscaling replace}{p_end}


{title:Also see}

{psee}
Manual:  {bf:[R] fracpoly}

{psee}
Online:  {help fracpoly postestimation};{break}
{helpb mfp}
{p_end}