{smcl}
{* 30june2008}{...}
{hline}
help for {hi:imputerasch}{right:Jean-Benoit Hardouin}
{hline}

{title:Imputation of missing binary variables by a Rasch model}

{p 8 14 2}{cmd:imputerasch} {it:varlist} [{cmd:,} {cmdab:pref:ix}({it:string}) {cmdab:noran:dom} {cmdab:savep:roba}({it:string}) {cmdab:nbit:eration}({it:#}) {cmdab:det:ails} {cmdab:max}({it:#})]

{p 8 14 2}{it:varlist} is a list of two or more existing dichotomous variables.

{title:Description}

{p 4 8 2}{cmd:imputerasch} imputes missing binary data by a Rasch model. 
The parameters of the Rasch model are estimated on complete data, then the missing data are imputed from the  estimated probability 
for each individual to response to each item.
By default, the imputed value is a result of a random draw within a Bernouilli random variable with this probability used like 
parameter, but it is possible to affect more deterministically the value of the missing data (0 if p<0.5 and 1 if p>=.5) with 
the {cmd:norandom} option.
 An iterative procedure can be run in a second time by estimating parameters of the Rasch model on existing and imputing data, 
 and by eventually correcting missing data at each step (see the {cmd:nbiteration} option).
 This procedure is stopped as soon the allowed maximal number of iterations is attained, or as soon the imputed values are stable.

{title:Options}

{p 4 8 2}{cmd:prefix}. The former variables (with missing data) are keeped. New variables are created by imputing new values to missing data. 
The name of these new variables are the names of the former variables preeceded by the prefix defined in this option. By default, this prefix is "imp".

{p 4 8 2}{cmd:norandom} avoids to randomly draw the value of imputation (by default). A deterministic process is used : if the expected probability is <0.5, imputed value is 0, else imputed value is 1 
(the old name of this option, {cmd:nobinomial}, continues to run).

{p 4 8 2}{cmd:saveproba} allows saving the expected probability in variables whose the names begin by the string defined in this option.

{p 4 8 2}{cmd:nbiteration} realizes an iterative procedure which is stopped as soon as the maximal number of iterations is attained, or as soon as the imputed data are stable.

{p 4 8 2}{cmd:details} gives details on the imputation.

{p 4 8 2}{cmd:max} allows imputing missing values only for individuals with a maximal number of missing values defined with this option.



{title:Example}

	{inp:. imputerasch item*}

	{inp:. imputerasch item*, norandom saveproba(p) prefix(dataimputed) max(4)}

	{inp:. imputerasch item1-item5, nbiteration(5) details}


{title:Author}

{p 4 8 2}Jean-Benoit Hardouin, PhD, assistant professor{p_end}
{p 4 8 2}EA 4275 "Biostatistics, Clinical Research and Subjective Measures in Health Sciences"{p_end}
{p 4 8 2}University of Nantes - Faculty of Pharmaceutical Sciences{p_end}
{p 4 8 2}1, rue Gaston Veil - BP 53508{p_end}
{p 4 8 2}44035 Nantes Cedex 1 - FRANCE{p_end}
{p 4 8 2}Email:
{browse "mailto:jean-benoit.hardouin@univ-nantes.fr":jean-benoit.hardouin@univ-nantes.fr}{p_end}
{p 4 8 2}Websites {browse "http://www.anaqol.org":AnaQol}
and {browse "http://www.freeirt.org":FreeIRT}