Question

我有一个日期（月），人和回报（每月）的数据集。我需要计算每个人从4月份t到3月份t + 1的数据集的复合月度回报。例如，

年度回报人A = 4月1年* 5月1 * * * 3月2年。

我可以知道在SAS中如何做到这一点？我需要一个阵列吗？

Answer 1

您的意思是您的数据集看起来像

日期，人物，回归

“01Apr2009”d，A，1.1

等

/* Then you can do */

data ur_input_data_view / view = ur_input_data_view;

  /* assuming your dataset is stored in ur_input_data */

  set ur_input_data;

  /* obtain the year and month from the date variable */

  year = year(date);

  mth = year(date);

run;

/* you can skip the sort if your data is already sorted by person then date */

proc sort data = ur_input_data;

  by person date;

run; 

data output_data;

  set ur_input_data_view ;

  by person date;

  /* reset annual return to null if looking at a new person or month is april */

  if first.person or mth = 4 then do;

    annual_return = return;

  end;

  retain annual_return;

  annual_return = annual_return * return;

  /* this will output nothing for the persons that don't have a record at march */

  if mth = 3 then output;

run;

计算SAS中的复合回报

1 个答案: