In cases where you are modeling under uncertainty, it is useful to have a random number generator to help simulate a situation. LINGO supports the @RAND function, which can be used to generate sequences of pseudo random numbers with uniform distribution between 0 and 1. By using the same seed value, you can recreate a series of numbers. In this example, we generate 15 random numbers, and then use them with the @PSN and @PTD functions to generate observations from both the unit Normal and T distributions.

MODEL:

! Generate a series of Normal and T distributed

 random variables ;

SETS:

 SERIES/1..15/: U, ZNORM, ZT;

ENDSETS

 

! First uniform is arbitrary;

 U( 1) = @RAND( .1234);

 

! Generate the rest recursively;

 @FOR( SERIES( I)| I #GT# 1:

  U( I) = @RAND( U( I - 1))

 );

 

! Generate some...;

 @FOR( SERIES( I):

!   Normal deviates...;

  @PSN( ZNORM( I)) = U( I);

!   and t deviates(2 degrees of freedom);

  @PTD( 2, ZT( I)) = U( I);

!   ZNORM and ZT may take on negative values;

  @FREE( ZNORM( I)); @FREE( ZT( I));

 );

END

Model: DEMRND