About HarshadMadhamshettiwar

HarshadMadhamshettiwar · ‎03-01-2015

Hello Experts, how do we create a plot using the set of data like below: Category element value date cat1 a 20 Jan-10 cat1 a 30 Feb-10 cat1 a 40 Mar-10 cat1 a 25 Apr-10 cat1 b 10 Jan-10 cat1 b 15 Feb-10 cat1 b 20 Mar-10 cat1 b 25 Apr-10 cat2 c 30 Jan-10 cat2 c 35 Feb-10 cat2 c 25 Mar-10 cat2 c 50 Apr-10 cat2 d 45 Jan-10 cat2 d 50 Feb-10 cat2 d 60 Mar-10 cat2 d 55 Apr-10 From above data we can see that , category "cat1" have two elements "a and b" , category "cat2" have two elements "c and d" and for all of them we have "values" and "date". Que is how do we create plot which looks like this: Motive is to see how category looks at an aggregated level along with how its elements are contributing. So above plot will be for "CAT1 and all its elements (here A and B)" and we will have same for "CAT2". Thanks, Harshad M.

HarshadMadhamshettiwar · ‎01-28-2015

Found here Sample 41732 /* Sample data */ data ds1; do date='01nov2010'd to '31dec2010'd; output; end; run; /* The WEEK variable returns a value of 1 to 5 based on the number of weeks in a month using a combination of the INTNX and INTCK functions. */ data ds2; set ds1; week=intck('week',intnx('month',date,0),date)+1; run; proc print; format date date9.; run;

HarshadMadhamshettiwar · ‎01-28-2015

From output of above solution how can we create Week of Month variable which will take only 5 values ('1','2','3','4','5') representing the first to fifth week of that month. This way week 5 of our dataset will be week 1 of February and WOM(week of month) variable should show "1" and respectively "2","3","4" for 6,7 and 8. Same way for March (WEEK) (WOM-new var) 9 1 10 2 11 3 12 4 13 5 Thanks. Harshad

HarshadMadhamshettiwar · ‎01-27-2015

Hello Experts, Now, I am in a situation where I have to use Hierarchical Cluster analysis but I am not being able to decide the number of clusters. I see Proc ACECLUS which says "Neither cluster membership nor the number of clusters needs to be known. PROC ACECLUS is useful for preprocessing data to be subsequently clustered by the CLUSTER or FASTCLUS procedure" But when I see the example provided (LONE example) in documentation section it uses "MAXC=3" option (which is offcourse mandatory requirement of FASTCLUS procedure and is like providing number of cluster explicitly - SAS/STAT(R) 9.2 User's Guide, Second Edition) if it is to be that way then what is the use of running ACECLUS when we are giving the number of clusters explicitly and why then it is quoted in above sentence number of cluster need not to be known. I am confused. Nevertheless main question is can we use FASTCLUS or CLUSTER procedure without Prior running ACECLUS (I think the answer is yes). But ACECLUS has got its own importance for calculating canonical variables if our dataset that have variables with different scalar measures. And if we use ACECLUS first, then how to arrive at desired number of clusters given that user is novice and is not aware of different algorithms and methods and business needs etc etc. Thanks. Harshad M.

HarshadMadhamshettiwar · ‎01-27-2015

Hi, I have a question here how did you land upon the number "10" even before running proc cluster, i.e. how did you decided to create 10 clusters. What was the motivation behind that, was a business requirement? Because otherwise pre-deciding the number of clusters in impossible and scientifically incorrect. Now, I am in a situation where I have to use Hierarchical Cluster analysis but I am not being able to decide the number of clusters. I see Proc ACECLUS which says "Neither cluster membership nor the number of clusters needs to be known. PROC ACECLUS is useful for preprocessing data to be subsequently clustered by the CLUSTER or FASTCLUS procedure" But when I see the example provided (LONE example) in documentation section it uses "MAXC=3" option (which is offcourse mandatory requirement of FASTCLUS procedure and is like providing number of cluster explicitly - SAS/STAT(R) 9.2 User's Guide, Second Edition) if it is to be that way then what is the use of running ACECLUS when we are giving the number of clusters explicitly and why then it is quoted in above sentence number of cluster need not to be known. I am confused. Nevertheless main question is can we use FASTCLUS or CLUSTER procedure without Prior running ACECLUS (I think the answer is yes). But ACECLUS has got its own importance for calculating canonical variables if our dataset that have variables with different scalar measures. And if we use ACECLUS first, then how to arrive at desired number of clusters given that user is novice and is not aware of different algorithms and methods and business needs etc etc. Thanks. Harshad M.

HarshadMadhamshettiwar · ‎01-27-2015

When I was trying to do it myself I used INTNX and MDY but I was not aware that we can use "WEEK-1"(variable in conjunction with mathematical operation) as parameter in INTNX. Thanks for answers.

HarshadMadhamshettiwar · ‎01-22-2015

Hi Experts, I have two variables "Year" and "Week" using these I need to create a SAS date. I know there is no standalone SAS function available which can do this task but using few functions in conjunction may be useful and tried few but failed. Here is my data year week 2014 1 2014 2 2014 3 2014 4 2014 5 2014 6 2014 7 2014 8 2014 9 2014 10 2014 11 2014 12 2014 13 2014 14 2014 15 2014 16 2014 17 2014 18 2014 19 2014 20 2014 21 2014 22 2014 23 2014 24 2014 25 2014 26 2014 27 2014 28 2014 29 2014 30 2014 31 2014 32 2014 33 2014 34 2014 35 2014 36 2014 37 2014 38 2014 39 2014 40 2014 41 2014 42 2014 43 2014 44 2014 45 2014 46 2014 47 2014 48 2014 49 2014 50 2014 51 2014 52 2014 53 And as we know week 1 in year 2014 ended on 4th - Jan- Saturday (SAS date Sunday=1 and Saturday=7). Similarly week 2 ended on 11th -January- Saturday and so on which will give me last value in dataset as 3rd-Jan-2015- Saturday(any format is OK). So to achieve this given the information and we have liberty to hard code 1st value as "Saturday 4 Jan 2014" if at all we need to use it for future reference in the date creation. Any thoughts. Thanks. Harshad M.

HarshadMadhamshettiwar · ‎12-29-2014

Oh yes it works, but can you also explain the logic behind the code. I mean first two IF conditions are not interdependent and if i am correct they are mutually exclusive. And code did not work with raw file/ Infile statement, why? How can we put this in simple English language for better understanding.

HarshadMadhamshettiwar · ‎12-29-2014

Yes..! Its working now and giving expected output. I know there may different ways of achieving the same result, Can you please explain not only how the code is working but also how to approach and think solutions for such data manipulations? Thanks

HarshadMadhamshettiwar · ‎12-29-2014

Hai.kuo , this code is showing error. Jagdishkatam, this seems not working correctly as values are truncating and its not starting "date" from first non zero instance for "name" "b".

HarshadMadhamshettiwar · ‎12-28-2014

name date value a Jan-11 0 a Feb-11 0 a Mar-11 0 a May-11 0 a Jun-11 0 a Jul-11 3 a Aug-11 15 a Oct-11 22 a Nov-11 21 a Dec-11 20 a Feb-12 0 a Mar-12 21 a Apr-12 0 b Jan-11 0 b Feb-11 0 b Mar-11 0 b May-11 22 b Jun-11 23 b Jul-11 55 b Aug-11 565 b Oct-11 566 b Nov-11 20 b Dec-11 20 b Feb-12 20 b Mar-12 11 b Apr-12 0 Experts, I am trying to read the data starting from first non-zero value for every "Name" here reading for "a" should start from July-11 and for "b" it should start from May-11. Can we do this by manipulating while reading raw file or we need have the data in sas dataset format and then only we use datasteps and procs to do task. Thanks.

HarshadMadhamshettiwar · ‎12-26-2014

Apology for that but it is partially true if we seek some datastep functions to the job. Nevertheless thanks everyone for motivating to participate and correcting newbies where we go wrong.

HarshadMadhamshettiwar · ‎12-26-2014

Thanks I used mix of both options.

HarshadMadhamshettiwar · ‎12-20-2014

Experts, Below is the output result of proc compare which compares two data sets and tells whether observations have matching values for variables. It the values are matching it gives the difference as 0 otherwise the calculated value. Now I want to limit this data to only those "C_name" for which all "_OBS_" are having "Value"=0 . Kindly see c_name "onn" is having one zero for its set of data and I dont want data relating to ONN to be in my new dataset. I only want those C_name for which all obs have 0 values; so the condition need to check whether the "value" is 0 for all the unique "C_names" . One more thing the _OBS_ reference may change i.e here I have posted only 6 but originally this number may be huge so its better if we have something which is dynamic and automatically COUNTS the number of _OBS_ for unique c_names which will be same for all c_name. _TYPE_ _OBS_ C_name metric value DIF 1 hero ar_y1 88 DIF 2 hero ar_y2 -22 DIF 3 hero ar_y3 -22 DIF 4 hero cash_y1 79 DIF 5 hero cash_y2 -10 DIF 6 hero cash_y3 -10 DIF 1 onn ar_y1 0 DIF 2 onn ar_y2 -20 DIF 3 onn ar_y3 -20 DIF 4 onn cash_y1 0 DIF 5 onn cash_y2 -22 DIF 6 onn cash_y3 75 DIF 1 yu ar_y1 0 DIF 2 yu ar_y2 0 DIF 3 yu ar_y3 0 DIF 4 yu cash_y1 0 DIF 5 yu cash_y2 0 DIF 6 yu cash_y3 0 Requirement in NUTSHELL: ALL the C_names which have all the values zero for all the obs (here we have only one C_name which fits our criteria and that is "YU"). Thanks a lot...!

HarshadMadhamshettiwar · ‎12-20-2014

Experts, Below is the output result of proc compare which compares two data sets and tells whether observations have matching values for variables. It the values are matching it gives the difference as 0 otherwise the calculated value. Now I want to limit this data to only those "C_name" for which all "_OBS_" are having "Value"=0 . Kindly see c_name "onn" is having one zero for its set of data and I dont want data relating to ONN to be in my new dataset. I only want those C_name for which all obs have 0 values; so the condition need to check whether the "value" is 0 for all the unique "C_names" . One more thing the _OBS_ reference may change i.e here I have posted only 6 but originally this number may be huge so its better if we have something which is dynamic and automatically COUNTS the number of _OBS_ for unique c_names which will be same for all c_name. _TYPE_ _OBS_ C_name metric value DIF 1 hero ar_y1 88 DIF 2 hero ar_y2 -22 DIF 3 hero ar_y3 -22 DIF 4 hero cash_y1 79 DIF 5 hero cash_y2 -10 DIF 6 hero cash_y3 -10 DIF 1 onn ar_y1 0 DIF 2 onn ar_y2 -20 DIF 3 onn ar_y3 -20 DIF 4 onn cash_y1 0 DIF 5 onn cash_y2 -22 DIF 6 onn cash_y3 75 DIF 1 yu ar_y1 0 DIF 2 yu ar_y2 0 DIF 3 yu ar_y3 0 DIF 4 yu cash_y1 0 DIF 5 yu cash_y2 0 DIF 6 yu cash_y3 0 Requirement in NUTSHELL: ALL the C_names which have all the values zero for all the obs (here we have only one C_name which fits our criteria and that is "YU"). Thanks a lot...!

Online Status	Offline
Date Last Visited	‎11-23-2016 05:30 AM

Tricky Sorting--How to decide and subset only first combination of var...

Re: How to make Proc Transpose scan all the rows-So that it does not s...

How to make Proc Transpose scan all the rows-So that it does not say n...

Re: Rename variables based on Mapping Table

Re: How to dynamically rename multiple variables

Re: Is it some kind of bug, Proc Import behaving incorrectly..!

Is it some kind of bug, Proc Import behaving incorrectly..!

Re: coding on forecast

Re: How we create only specific row(observations)/How do we create row...

Re: How we create only specific row(observations)/How do we create row...

Re: How to make Proc Transpose scan all the rows-So that it does not s...

Re: How to make Proc Transpose scan all the rows-So that it does not s...

Re: How we create only specific row(observations)/How do we create row...

Re: How we create only specific row(observations)/How do we create row...

Re: How we create only specific row(observations)/How do we create row...

Re: How to dynamically rename multiple variables

How to dynamically rename multiple variables

Re: Rename variables based on Mapping Table

Re: WOW I didn't know SAS does not have Function for this...Do YOU hav...

How to create a GRAPH like THIS(plotting a category and its constituen...

Re: How do we create SAS Date from Year and Week information only.?

Re: How do we create SAS Date from Year and Week information only.?

(Magic)Number of Clusters....How do we arrive at desired number of clu...

Re: Exogenously imposing the number of clusters in Proc Cluster

Re: How do we create SAS Date from Year and Week information only.?

How do we create SAS Date from Year and Week information only.?

Re: How to ignore leading zeros or read data from first non-zero value...

Re: How to ignore leading zeros or read data from first non-zero value...

Re: How to ignore leading zeros or read data from first non-zero value...

How to ignore leading zeros or read data from first non-zero value in ...

Re: WOW I didn't know SAS does not have Function for this...Do YOU hav...

Re: WOW I didn't know SAS does not have Function for this...Do YOU hav...

WOW I didn't know SAS does not have Function for this...Do YOU have an...

WOW I didn't know SAS does not have Function for this...Do YOU have an...