SAS中文论坛

标题: 请问怎样转换这个数据的格式？ [打印本页]

作者: shiyiming 时间: 2004-5-25 03:38
标题: 请问怎样转换这个数据的格式？
比如时间数列格式为；

YEAR  JAN FEB MAR。。。。
1992 89 34 56
1993 90 98 76
1994 32 56 87

。。。。。
转换成
JAN1992  89
FEB1992  34
MAR1992 56
。。。。。。。

谢谢了！

作者: shiyiming 时间: 2004-5-25 10:21
data jg;
  set ex;/*原始数据*/
  array data _numeric_;
  do i=1 to 12;
date=mdy(i,1,data{1});
num=data{i+1};
output;
  end;
  format date monyy7.;
  keep date num;
run;
其中mdy函数将月日年转为日期，我没找到仅仅转为月年的函数

作者: shiyiming 时间: 2004-5-25 21:32
非常感谢GBT朋友，我先试试。

作者: shiyiming 时间: 2004-5-25 22:11
Obs          year jan       feb       mar       apr       may       jun

1          1992 140,555    141,037    142,205    142,469    141,139    143,070
2          1993 150,578    149,325    146,865    150,091    152,720    149,890
3          1994 153,649    154,307    157,063    156,658    157,825    159,359
4          1995 171,989    173,509    171,765    174,247    174,973    176,509
5          1996 180,514    181,534    183,910    183,751    184,293    185,330
6          1997 190,891    195,484    194,038    192,815    194,107    194,091
7          1998 197,373    197,332    198,588    198,442    198,398    197,191
8          1999 200,092    202,494    204,619    205,573    210,072    211,370
9          2000 224,266    223,581    226,181    227,475    227,656    230,653
  10          2001 231,919    231,106    227,998    226,514    227,599    223,126
  11          2002 221,219    223,009    223,315    226,696    227,493    228,840
  12          2003 235,248    235,557    237,727    234,998    234,073    237,209
  13          2004 253,664    259,109    266,213

Obs jul       aug       sep       oct       nov       dec

1 146,114    144,479    146,259    146,181    146,552    145,554
2 151,342    151,798    152,033    152,549    152,560    149,960
3 160,764    165,445    164,606    167,360    167,970    170,352
4 176,413    177,577    178,342    179,075    181,040    183,607
5 186,883    188,271    188,101    189,382    190,788    188,783
6 194,118    194,000    196,804    195,953    196,059    195,203
7 198,259    197,049    197,780    199,016    199,066    200,492
8 212,417    214,001    215,417    218,591    220,626    222,130
9 229,532    229,118    231,415    232,197    230,377    233,327
  10 224,580    224,778    224,550    219,797    220,125    219,506
  11 229,345    232,391    231,790    231,287    233,819    232,033
  12 239,113    240,029    241,527    246,430    247,737    251,526
  13

以上是SAS打印出的文件，运行程序后系统报错为：

ERROR: Array subscript out of range at line 62 column 5.
year=1992 jan=140,555 feb=141,037 mar=142,205 apr=142,469 may=141,139 jun=143,070 jul=146,114
aug=144,479 sep=146,259 oct=146,181 nov=146,552 dec=145,554 _I_=. i=1 date=JAN1992 num=. _ERROR_=1
_N_=1

请问是什么原因？谢谢了

作者: shiyiming 时间: 2004-5-25 23:13
Plese check your EX data set to see if the variable JAN to DEC are numeric, if they are character variables, the whole thing will be screwed up.

作者: shiyiming 时间: 2004-5-27 21:54
谢谢XIC，SAS中象JAN，这样的字符可以定义为数值变量吗？怎么定义呢？不行的话，就把他改成 1 ，2， 3算了。谢谢了。

作者: shiyiming 时间: 2004-5-28 08:59
Wow, it will be difficult to explain from the very beginning. I was talking about JAN to DEC as variable names, not the values. The value of 1, 2, 3 can also be characters, and the value as JAN, FEB and so on can also be numeric if you attach a format on it. I would like you to check the variables JAN to DEC to see if they are character variables. If they are, the program posted by GBT may need a minor modification.

作者: shiyiming 时间: 2004-5-28 09:21
data ex;
  input year jan$8. feb$8. mar$8. apr$8. may$8. jun$8. jul$8. aug$8. sep$8. oct$8. nov$8. dec$8. ;
dec$ ;
  cards;
1992 140,555 141,037 142,205 142,469 141,139 143,070  146,114 144,479 146,259 146,181 146,552 145,554
1993 150,578 149,325 146,865 150,091 152,720 149,890 151,342 151,798 152,033 152,549 152,560 149,960
1994 153,649 154,307 157,063 156,658 157,825 159,359 160,764 165,445 164,606 167,360 167,970 170,352
1995 171,989 173,509 171,765 174,247 174,973 176,509 176,413 177,577 178,342 179,075 181,040 183,607
1996 180,514 181,534 183,910 183,751 184,293 185,330  186,883 188,271 188,101 189,382 190,788 188,783
1997 190,891 195,484 194,038 192,815 194,107 194,091 194,118 194,000 196,804 195,953 196,059 195,203
1998 197,373 197,332 198,588 198,442 198,398 197,191 198,259 197,049 197,780 199,016 199,066 200,492
1999 200,092 202,494 204,619 205,573 210,072 211,370 212,417 214,001 215,417 218,591 220,626 222,130
2000 224,266 223,581 226,181 227,475 227,656 230,653 229,532 229,118 231,415 232,197 230,377 233,327
2001 231,919 231,106 227,998 226,514 227,599 223,126 224,580 224,778 224,550 219,797 220,125 219,506
2002 221,219 223,009 223,315 226,696 227,493 228,840 229,345 232,391 231,790 231,287 233,819 232,033
2003 235,248 235,557 237,727 234,998 234,073 237,209 239,113 240,029 241,527 246,430 247,737 251,526
2004 253,664 259,109 266,213
;
data jg;
set ex;
array y _numeric_ ;
array data _character_;
do i=1 to 12;
date=mdy(i,1,y{1});
num=data{i};
output;
end;
format date monyy7.;
keep date num;
run;

应该没问题了

作者: shiyiming 时间: 2004-6-8 23:19
标题: thanks
几天没有上网，感谢大家的帮助。

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