help centcalc                                                   Patrick Royston
-------------------------------------------------------------------------------

Centile calculation

Syntax

centcalc mvar svar [, options ]

options Description ------------------------------------------------------------------------- centiles(numlist) defines the required centiles of yvar|xvar dist(dist_name) specifies the distribution of the response variable gamma(# | gvar) G delta(# | dvar) D cv produces estimated parameters and centiles at all non-missing values of xvar prefix(stub) defines a censoring variable -------------------------------------------------------------------------

Description

centcalc calculates distribution-based centiles from user-supplied parameter values for location (mvar), scale (svar) and shape (gamma() and delta()). The distributions (dist()) are all transformations of the Normal (Gaussian) distribution.

Options

centiles(numlist) is the list of estimated centiles (e.g. 50 for the median) to be calculated.

cv is an option relating to the parameterisation of the S-curve for the chosen distribution. See xriml for more information.

delta(# | dvar) defines the (second) shape parameter as a variable (dvar) or as a number (#). If delta() is not specified, it is taken from the macro $S_delta.

dist(dist_name) determines the distributional model (see xriml and Remarks below for further details). dist_name must be one of n, en, men, pn, mpn, or sl. If dist() is not specified, dist_name is taken from the macro $S_dist.

gamma(# | gvar) defines the (first) shape parameter as a variable (gvar) or as a number (#). If gamma() is not specified, it is taken from the macro $S_gamma.

prefix(stub) is the prefix of the names of new variables created to hold the estimated centiles, each labelled appropriately. Default stub is C.

Remarks

Further details of the distributions:

-------------------------------------------------------------------------- Code Name of distribution Location Scale Shape param(s) -------------------------------------------------------------------------- n Normal (Gaussian) mean std. dev. (not applic.) en Exponential-Normal median std. dev.-like skewness-like men Modulus exp-Normal median std. dev.-like (see below) pn Power-Normal median std. dev.-like power transf. mpn Modulus pow-Normal median std. dev.-like (see below) sl Shifted lognormal median std. dev.-like skewness-like --------------------------------------------------------------------------

The mpn and men distributions each have two shape parameters, the first (denoted by G) related to skewness and the second (denoted by D) related to kurtosis.

Examples

. centcalc M_ml S_ml, dist(n) centiles(10 90)

. centcalc M_ml S_ml, dist(en) centiles(10 90) gamma(G_ml)

. centcalc median cv, dist(pn) centiles(3 25 75 97) gamma(lambda) cv

. centcalc M S, dist(men) centiles(1 3 50 97 99) gamma(g) delta(d)

Author

Patrick Royston, MRC Clinical Trials Unit, London. patrick.royston@ctu.mrc.ac.uk

Eileen Wright, Macclesfield

Also see

Manual: [R] fracpoly

Online: xriml, xrigls (if installed), fracpoly