Hi,    I want to change a certain value of prior period that is very close to one of the following periods.    For example,   Data want; input Firm Year Change Solved; datalines; 1001 2000 0 0 1001 2001 0 0 1001 2002 1 0 1001 2003 0 0 1001 2004 1 1 1001 2005 0 0 1001 2006 0 0 1002 2002 0 0 1002 2003 1 0 1002 2004 1 0 1002 2005 1 1 1002 2006 0 0 1002 2007 0 0 1003 1999 0 0 1003 2000 1 1 1003 2001 0 0 1003 2002 0 0 1003 2003 0 0 1003 2004 1 1 1003 2005 0 0 ; run;   The solved column is what I want to have in my dataset.   If a firm has a change = 1 in very close period (within 3 years) with the next period when change = 1, I want to make the former change = 1 to 0. That is, I want to have at least 3 years gap between changes by firms.   Please help me this problem.   Thank you! ... View more

Hi,   Could you please help me extract first name of full name?   Data test; input id fullname $; 1 Michael R. Boyce 2 James G. Brocksmith, Jr. 3 Gerald F. Fitzgerald, Jr. 4 Norman R. Bobins, BS, MBA 5 Peter Pace, USMC, Ret. 6 Ryan, Norman A. 7 Anne Newman Foreman ; run;   I need to create "first_name" column with middle name and without middle name and "last_name" using the fullname information. As you can see above, In ID=6, the full name start with the last name unlike other observations. If it is the case, I need to re-order it to be consistent with others. Like...   Data test; set test; input ID First_name_with_middle $ Last_name $ First_name_without_middle $  datalines; 1 Michael R. Boyce Michael 2 James G. Brocksmith James 3 Gerald F. Fitzgerald Gerald 4 Norman R. Bobins Norman 5 Peter Pace Peter  6 Norman A. Ryan Norman 7 Anne Newman Foreman Anne  ; run;     That is, the first name with middle name column should contain the middle name abbreviation as well. And, the last name column only contains last name of the full name. Please help me!   ... View more

Hi,   Could you help me figure out how to calculate percentage by firm and year?   My data looks like...    data test; input firm year director dummy; datalines; 1001 2000 aaa 0 1001 2000 bbb 0 1001 2000 ccc 1 1001 2000 ddd 0 1001 2001 aaa 0 1001 2001 bbb 0  1001 2001 ccc 1 1001 2001 ddd 0 1001 2001 eee 0 1002 2005 fff 1 1002 2005 ggg 1 1002 2005 hhh 0 1002 2005 iii 0 1002 2006 fff 1 1002 2006 ggg 0 1002 2006 hhh 1 1002 2006 iii 0 1002 2006 jjj 1 ; run;   I need to get a column (Percentage_MD) having the percentage of MD by firm and year.   Thank you! ... View more

Hi,   I need to retain only -3 and +3 period observations for each group. Now my data looks like...   Data example; input firm manager $ year turnover post_turnover turnover_year; datalines; 1001 abc 2005 0 0 . 1001 abc 2006 0 0 . 1001 abc 2007 1 1 2007 1001 abc 2008 0 0 . 1001 abc 2008 0 1 . 1001 abc 2009 0 0 . 1001 abc 2009 0 1 .  1001 abc 2010 0 0 .  1001 abc 2010 0 1 . 1001 abc 2011 0 0 .  1001 abc 2011 0 1 . 1001 abc 2012 1 1 2012 1001 abc 2013 0 1 . 1002 def 1990 0 0 . 1002 def 1991 0 0 . 1002 def 1992 0 0 . 1002 def 1993 0 0 . 1002 def 1994 1 1 1994 1002 def 1995 0 0 . 1002 def 1995 0 1 .   1002 def 1996 1 1 1996 1002 def 1996 0 1 . ; run;   As you can see, I have multiple turnovers by each firm. So, observations between two turnovers were duplicated because they can be both post and pre.  Now my problem is how I can retain only observations within three years around turnover_year (i.e., -3 and +3) for each firm. Also, I need to exclude the observations where the turnover was occurred.   Please help me this problem.   Thank you SAS masters!    ... View more

Hi SAS masters,   I am trying to clean my dataset for -3 and +3 year window.   My current data looks like..   Data test; input ID MANAGER YEAR POST_TURNOVER; datalines; 1001 a 1990 0 1001 a 1991 0 1001 a 1992 0 1001 a 1993 0 1001 a 1994 0 1001 a 1995 1 1001 a 1996 1 1001 a 1997 1 1001 a 1998 1 1001 a 1999 1 1001 a 1995 0 1001 a 1996 0 1001 a 1997 0 1001 a 1998 0 1001 a 1999 0 1001 a 2000 0 1001 b 2000 1 1001 b 2001 1 1001 b 2002 1 1001 b 2003 1 1002 c 1989 0 1002 c 1990 0 1002 c 1991 0 1002 c 1992 0 1002 c 1993 1 1002 c 1994 1 1002 c 1995 1 1002 c 1996 1 1002 c 1997 1  ; run;   My question here is how I can specify my dataset with -3 and +3 window (excluding the turnover year) for turnovers by ID and Manager. My data should look like..   ID MANAGER YEAR POST_TURNOVER Turnover       1001 a 1992 0 0 1001 a 1993 0 0 1001 a 1994 0 0 1001 a 1995 1 1 1001 a 1996 1 0 1001 a 1997 1 0 1001 a 1998 1 0 1001 a 1998 0 0 1001 a 1999 0 0 1001 a 2000 0 0 1001 b 2000 1 1 1001 b 2001 1 0 1001 b 2002 1 0 1001 b 2003 1 0 1002 c 1990 0 0 1002 c 1991 0 0 1002 c 1992 0 0 1002 c 1993 1 1 1002 c 1994 1 0 1002 c 1995 1 0 1002 c 1996 1 0 Thank you! ... View more