SAS中文论坛

标题: A SAS MACRO FOR DECISION STUMP [打印本页]

作者: shiyiming 时间: 2010-10-22 13:24
标题: A SAS MACRO FOR DECISION STUMP
From Wensui Liu's blog

*************************************************************; * A SAS MACRO FOR DECISION STUMP SIMILAR TO %SPLIT() MACRO *; * IN "Pharmaceutical Statistics Using SAS" by Dmitrienko, *; * Chuang-Stein, AND D'Agostino *; * --------------------------------------------------------- *; * GENERAL CONCEPT: *; * 1. DECISION STUMP IS A NAIVELY SIMPLE 1-LEVEL DECISION *; * TREE WITH TWO TERMINAL NODES *; * 2. COMMMONLY USED AS COMPONENTS IN BAGGING / BOOSTING *; * MACHINE LEARNING ENSEMBLE ALGORITHMS *; * --------------------------------------------------------- *; * PRACTICAL USAGES: *; * 1. VARIABLES PRE-SCREENING BEFORE MODEL DEVELOPMENT *; * 2. POINT SEARCHING FOR SCORECARD CUTOFF STRATEGY *; * --------------------------------------------------------- *; * AUTHOR: <a href="mailto:wensliu@paypal.com">wensliu@paypal.com</a> *; *************************************************************; data example1; do i = 1 to 1000; * x1: key driver with single cutoff = 5 *; x1 = 10 * ranuni(1); * x2: related with 2 different cutoffs *; x2 = 10 * ranuni(2); * x3: unrelated *; x3 = 10 * ranuni(3); if (x1 < 5 and x2 < 1.5) then y = 0; if (x1 < 5 and x2 >= 1.5) then y = 1; if (x1 >= 5 and x2 < 7.5) then y = 0; if (x1 >= 5 and x2 >= 7.5) then y = 1; w = 1; output; end; run; %macro stump(data = , w = , y = , xlist = ); %let i = 1; %local i; proc sql; create table _out ( variable char(32), gt_value num, gini num ); quit; %do %while (%scan(&xlist, &i) ne %str()); %let x = %scan(&xlist, &i); data _tmp1(keep = &w &y &x); set &data; where &y in (0, 1); run; proc sql; create table _tmp2 as select b.&x as gt_value, sum(case when a.&x <= b.&x then &w * &y else 0 end) / sum(case when a.&x <= b.&x then &w else 0 end) as p1_1, sum(case when a.&x > b.&x then &w * &y else 0 end) / sum(case when a.&x > b.&x then &w else 0 end) as p1_2, sum(case when a.&x <= b.&x then 1 else 0 end) / count(*) as ppn1, sum(case when a.&x > b.&x then 1 else 0 end) / count(*) as ppn2, 2 * calculated p1_1 * (1 - calculated p1_1) * calculated ppn1 + 2 * calculated p1_2 * (1 - calculated p1_2) * calculated ppn2 as gini from _tmp1 as a, (select distinct &x from _tmp1) as b group by b.&x; insert into _out select "&x", gt_value, gini from _tmp2 having gini = min(gini); drop table _tmp1; quit; %let i = %eval(&i + 1); %end; proc sort data = _out; by gini; run; proc report data = _out box spacing = 1 split = "*" nowd; column("DECISION STUMP SUMMARY" variable gt_value gini); define variable / "VARIABLE" width = 30 center; define gt_value / "CUTOFF VALUE*(GREATER THAN)" width = 15 center; define gini / "GINI" width = 10 center format = 9.4; run; %mend stump; %stump(data = example1, w = w, y = y, xlist = x1 x2 x3); /* ----------------------------------------------------------- | DECISION STUMP SUMMARY | | CUTOFF VALUE | | VARIABLE (GREATER THAN) GINI | |---------------------------------------------------------| | x1 | 4.9742638 | 0.3125 | |---------------------------------------------------------| | x2 | 7.4602286 | 0.3534 | |---------------------------------------------------------| | x3 | 0.6173268 | 0.4939 | ----------------------------------------------------------- */

欢迎光临 SAS中文论坛 (https://mysas.net/forum/)