assign locf and

[thanikondharish]

Name	Sex	Age	Height	Weight
Alfred	M	14	69	112.5
Alice	F	13	56.5	84
Barbara	F	13	65.3	98
Carol	F	14	62.8	102.5
Henry	M	14	63.5	102.5
James	M	12	57.3	.
Jane	F	12	59.8	.
Janet	F	15	62.5	112.5
Jeffrey	M	13	62.5	84
John	M	12	59	.
Joyce	F	11	51.3	50.5
Judy	F	14	64.3	90
Louise	F	12	56.3	.
Mary	F	15	66.5	112
Philip	M	16	72	150
Robert	M	12	64.8	.
Ronald	M	15	67	133
Thomas	M	11	57.5	85
William	M	15	66.5	112

in above data weight variable has some missings so we use locf method and fill up missing values and the same time assign dtype=locf like below table

new_weight	Name	Sex	Age	Height	Weight	dtype
112.5	Alfred	M	14	69	112.5
84	Alice	F	13	56.5	84
98	Barbara	F	13	65.3	98
102.5	Carol	F	14	62.8	102.5
102.5	Henry	M	14	63.5	102.5
102.5	James	M	12	57.3	.	locf
102.5	Jane	F	12	59.8	.	locf
112.5	Janet	F	15	62.5	112.5
84	Jeffrey	M	13	62.5	84
84	John	M	12	59	.	locf
50.5	Joyce	F	11	51.3	50.5
90	Judy	F	14	64.3	90
90	Louise	F	12	56.3	.	locf
112	Mary	F	15	66.5	112
150	Philip	M	16	72	150
150	Robert	M	12	64.8	.	locf
133	Ronald	M	15	67	133
85	Thomas	M	11	57.5	85
112	William	M	15	66.5	112

[RW9]

Please post test data in the form of a datastep and clearly describe what it is you are asking.  From what you have posted:

data want;
  set have;
  dtype="locf";
run;

Gets you the second table, although this isn't what you are wanting is it.  Maybe something like:

data want (drop=weight rename=(new_weight=weight));
   set have;
   retain new_weight;
   if weight ne . then new_weight=weight;
   else dtype="locf";
run;
     

Not tested as no test data.

[thanikondharish]

use then do

[RW9]

?

[ballardw]

Are you asking how to implement locf?

[error_prone]

The paper http://www2.sas.com/proceedings/sugi28/099-28.pdf seems to contain a description on how to implement locf.