Proc mi technical question

[ndang3188]

When I tried to run proc mi for the following imputation, I noticed that the order of variables in regression model and in the var statement affects the imputed values. I want to understand what 

proc mi data=WIDEDAT seed=93538001 nimpute=50 out=OUTMI noprint

  minimum=0 maximum=14 minmaxiter=600 ;

  by TRT01PN;

  class CSIRSTR ;

  fcs nbiter=50 reg(BASE = CSIRSTR);

  fcs nbiter=50 reg(W2 = CSIRSTR BASE); /** Imputing missing mDSD values at W2**/

  fcs nbiter=50 reg(W4 = CSIRSTR BASE W2); /**Imputing missing mDSD values at W4 **/

  fcs nbiter=50 reg(W8 = CSIRSTR BASE W2 W4); /**Imputing missing mDSD values at W8 **/

  fcs nbiter=50 reg(W12 = CSIRSTR BASE W2 W4 W8);  /** Imputing missing mDSD values at W12 **/

  fcs nbiter=50 reg(W16 = CSIRSTR BASE W2 W4 W8 W12); /** Imputing missing mDSD values at W16 **/

  fcs nbiter=50 reg(W20 = CSIRSTR BASE W2 W4 W8 W12 W16);    /** Imputing missing mDSD values at W20 **/ 

  fcs nbiter=50 reg(W24 = CSIRSTR BASE W2 W4 W8 W12 W16 W20);    /** Imputing missing mDSD values at W24 **/ 

  var CSIRSTR BASE W2 W4 W8 W12 W16 W20 W24;

run;

[SAS_Rob]

This is to be expected.  Because the imputed values are a result of random draws from their respective distributions, changing the ordering of the variables will change where in the stream each distribution is being drawn from.  What should not change however is the posterior distribution from which those random numbers are drawn.  Think of it as similar to changing the seed.  You will get the same posterior distribution, but the draws from those distributions will differ.