I am a statistics student.  For this project I will have to use WEKA software but I thought it was a good opportunity to learn some new sas as well.  I'll be working in BASE SAS, SAS/STAT.    I have a dataset with 130 college courses over 5 years and 448 students.   I would like to find concentrations of classes in groups of 3 or 4 to recommend concentrations.    I'm looking for some ideas to start and I will continue the research on my own.  Is K-Means the right approach for something with this many values?   Cheers,   Mark ... View more

Below is the code I'm using, I will attach a work doc with the output.  I'm trying to order the graph on the x axis.  I tried a variable named sort but I'd like to have the dates as labels and I couldn't get the output to look the way I wanted.  I added a number to preface the dates, it's working but I don't like the way it looks.  So if there is an easy way to sort the x axis without labeling what the sort order is that would be a big help.  Also notice the height of the bars do not coincide with the hight of the labels.  When I started this I thought both would be an easy fix but I can't find anything.   I also tried formating the dates as date9. but the values changed in the gchart step.  Here's my code:   proc freq data=mdj._1l_threshold_report_&monyear noprint;table 'Pass/Fail'n / out=Current; proc freq data=mdj._1l_threshold_report_&p_monyear noprint;table pass_fail / out=previous; proc freq data=mdj._1l_threshold_report_&p_monyear2 noprint;table pass_fail / out=previous2; goptions reset=all cback=white border          htitle=11pt htext=11pt; proc sql; create table graph_prep as select catx(' ','3)',"&Current") as Date ,count,'Pass/Fail'n from Current union select catx(' ','2)',"&previous") as Date ,count,pass_fail as 'Pass/Fail'n from previous union select catx(' ','1)',"&previous2") as Date ,count,pass_fail as 'Pass/Fail'n from previous2 order by date; data HE_Graph;    length function color text $ 8 style $ 20;    retain function 'label' color 'black' when 'a'           xsys ysys '2' position 'E' size 3 hsys '3';    set graph_prep;    style="'Albany AMT'";    midpoint=date;    subgroup='Pass/Fail'n;    text = count; /*   text=left(put(sales,dollar8.));*/ run;  /* Define axis characteristics */ axis1 minor=none;* label=(count);  /* Add a title to the graph */ title 'Executive Summary';    /* Produce the bar chart using the ANNO= */  /* option on the VBAR statement.         */ proc gchart data=graph_prep;    vbar date / sumvar=count raxis=axis1                    nozero anno=HE_Graph subgroup='Pass/Fail'n                width=12 space=3; run; quit;                   ... View more

I am manipulating code I get from using the Enterprise Import feature.  The original code is this:       INFILE '/sas/saswork/SAS_work39CA0000252B_scsmt300a/#LN00017'         LRECL=467         ENCODING="LATIN1"         TERMSTR=CRLF         DLM='7F'x         MISSOVER         DSD ;     INPUT         sys              : $CHAR6.         LOAN_NUMBER      : ?? BEST20.         LOAN_CLOSING_DATE : ?? YYMMDD8.         FIRST_PAYMENT_DATE : ?? YYMMDD8.   I changed the code to allow for the maximum lengths that the Federal Reserve may provide, I also changed the date to character format.  Here is what I have:   INFILE "/sas/user_data/general/sandbox/pp78499/&filename"         firstobs=2         LRECL=542 /*        ENCODING="LATIN1"*/ /*        TERMSTR=crlf*/         DLM='|'         MISSOVER         DSD ;     INPUT         sys              $CHAR6.         LOAN_NUMBER      $char33.         LOAN_CLOSING_DATE $char10.         FIRST_PAYMENT_DATE $char10.   an example of the output is below:   ACLShe    |0000000000000000|20040716|200408    15|PA|1908    2|15000|15 ACLShe    |0000000000000000|19990604|199907    10|PA|1932    0|176663|1 ACLShe    |0000000000000000|20010213|200104    05|PA|1913    6|40000|40 ACLShe    |0000000000000000|20010215|200103    20|PA|1903    8|104000|1   The pipes are coming into the output, not deliminating the data.  Any help will be appreciated.   Thank You,   Mark         ... View more

I have a large number of edits/rules that are provided by the federal reserve that I run against our mortgage data.  Several (38) sub totals are required to get the correct percentage.  In order to label these different threshold I added a new column for each of the 38 like so:   if bankruptcy_flag = 'Y' then do; _df4 = "df4"; '_54_Bankruptcy Loans Only'n = bankruptcy_chapter not in('7','9','11','12','13','U'); end; if investor_type = '7' then do; _df5="df5"; if com_loan_flg ne 1 and paid_in_full_coding = 0 and missing(baselII_pd) or    com_loan_flg in (1,.) and paid_in_full_coding = 0 and not missing(baselII_pd) and baselII_pd<.0003 or    baselII_pd>1 then '_125.1_Basel II BHC Portfolio'n = 1;   Later in the process I join each subtotal like this:   select _df1 as _df,count(loan_number) as Count from First_Lien_Rules_ETL_&propdate where _df1 = "df1" union select _df2 as _df,count(loan_number) as count from First_Lien_Rules_ETL_&propdate where _df2 = 'df2' union select _df3 as _df,count(loan_number) as count from First_Lien_Rules_ETL_&propdate where _df3 = 'df3' union   the _df field will give me the correct total, so I can look at the amount of fails/subtotal = percentage fail.    Within my experience this is the best solution I came up with.  The entire process takes under 30 minutes but this step takes over 10.  I'd love to get that time down.  Can anyone give me suggestions for another solution to the same output?   Thank You,     ... View more

I am using the following code.  Please notice the drop= _d: and keep= _:   There are a large amount of columns that are affected by this, this step generates a table with 1999410 observations and 465 variables, that later needs to be sorted.  It is by far the slowest part of my process.  Write now I'm using excel, copy/paste to write out a sql union statement with an order by after, I'm going to test the performance of that in comparison. I am wondering if there is a better way.     data first_lien_combine_&monyear(drop=_d:); set First_Lien_Rules_ETL_&propdate (keep= sys loan_number Run_Date reporting_month _: )     cp_rules_&monyear (keep = sys loan_number Run_Date reporting_month _: ); run;   Thank You in Advance,   Mark quit;   ... View more