相邻年间转移次数的统计

yangz98 发表于 2013-11-2 22:54
跑了一遍貌似答案不太对。。。
我也一木有想明白，lag(site)*site是什么意思？

lag是指滞后一期嘛  所以就是前一项啊

邓贵大 发表于 2013-11-2 12:25
Sorry, I didn't test my code.
Change 'if eof' to 'if _n_=12' or use a second data step. Didn't real ...

修改后有跑了几次，发现结果不对。。。

yangz98 发表于 2013-11-4 12:39
修改后有跑了几次，发现结果不对。。。

Dude, now I kind regret that I had wanted your easy bucks in the first place. Isn't God's love enough for me? Why would I want these fake Lun Tan Bi? There's something wrong in my soul. I got to quit for a few months.

1. data quextion;
   
2.         infile datalines eof=last;
   
3.         input year site;
   
4.         array x[5,5] _temporary_;
   
5.         array a[5];
   
6.         site_1=lag(site);
   
7.         if _n_=1 then call pokelong(repeat(put(0,rb8.), 25), addrlong(x[1,1]), 25*8);
   
8.         else x[site_1, site] = x[site_1, site]+1;
   
9.         last:
   
10.                 do i=1 to 5;
    
11.                         do j=1 to 5;
    
12.                                 a[j] = x[i,j];
    
13.                         end;
    
14.                         output;
    
15.                 end;
    
16.         keep a1-a5;
    
17. cards;
    
18. 1980 5
    
19. 1981 1
    
20. 1982 2
    
21. 1983 2
    
22. 1984 1
    
23. 1985 2
    
24. 1986 1
    
25. 1987 3
    
26. 1988 1
    
27. 1989 3
    
28. 1990 2
    
29. 1991 1
    
30. 1992 3
    
31. 1993 3
    
32. 1994 4
    
33. 1995 4
    
34. 1996 5
    
35. 1997 5
    
36. 1998 5
    
37. 1999 4
    
38. 2000 4
    
39. 2001 1
    
40. ;

复制代码

1. data quextion;
   
2.         infile datalines;
   
3.         input year site;
   
4. cards;
   
5. 1980 5
   
6. 1981 1
   
7. 1982 2
   
8. 1983 2
   
9. 1984 1
   
10. 1985 2
    
11. 1986 1
    
12. 1987 3
    
13. 1988 1
    
14. 1989 3
    
15. 1990 2
    
16. 1991 1
    
17. 1992 3
    
18. 1993 3
    
19. 1994 4
    
20. 1995 4
    
21. 1996 5
    
22. 1997 5
    
23. 1998 5
    
24. 1999 4
    
25. 2000 4
    
26. 2001 1
    
27. ;
    
28. data q/view=q;
    
29.         set quextion;
    
30.         site_1= lag(site);
    
31.         keep site_1 site;
    
32. proc means data=q nway noprint completetypes;
    
33.         class site_1 site;
    
34.         output out=q1;
    
35. proc transpose data=q1 out=anxwer(drop=_name_ _label_) prefix=a;
    
36.         by site_1;
    
37.         id site;
    
38.         var _freq_;
    
39. run;

复制代码

Final word, Dywane, you're doing a heck great of job on the basketball court. I wish you success in SAS too.