About SeanZ

SeanZ · ‎02-06-2014

Thank you for your help. This works when I only have two dataset. Now the problem would be when I have more than two columns to merge and I use a loop to emerge them. When I merge dataset1 and dataset2 I got correct dataset with firmId that combines those of dataset1 and dataset2. When I tried to merge dataset with dataset3, the column firmId was replaced and there are still missing firmId. How to solve this?

SeanZ · ‎02-06-2014

I have a ten dataset with columns firmid and size. I want to full join them via firmid. I wrote the following code, but give me undesirable results. Please see below. dataset 1 firmid size 1 100 2 200 3 250 dataset 2 firmid size 2 210 4 150 I want to get desired firmid size1 size2 1 100 . 2 200 210 3 250 . 4 . 150 But with the following codes proc sql; create table desired as select * from dataset1(rename=(size=size1)) full join dataset2(rename=(size=size2)) on dataset1.firmid = dataset2.firmid; quit; I got firmid size1 size2 1 100 . 2 200 210 3 250 . . . 150 Please help.

SeanZ · ‎02-06-2014

I coded a macro loop to get a value size for each month(loop), and then want to merge all size for every month into one dataset, but I have difficulty to achieve this purpose. Let me illustrate my attempt.For each loop I will get one dataset containing firmid and size, but for each loop the firmid may be different (most of them are overlapped). For instance, month 1 would be like firmid size 1 100 2 150 5 130 month 2 would be something like firmid size 1 120 2 140 4 100 What I want to acheive is something like firmid size(m1) size(m2) 1 100 120 2 150 140 4 . 100 5 130 . Please help. Thanks.

SeanZ · ‎02-03-2014

I have a code that I can execute in my local machine without any problem. Now I want to submit my code to the remote server for executing. I need to make a some adjustments. In my code I have many macro variables defined using %let. As well as some dataset in my local machine. To submit the code to remote server, I have the following paragraph. %let wrds=wrds.wharton.upenn.edu 4016; options comamid=TCP remote=WRDS; signon username= xxxxxx; rsubmit; some codes endrsubmit; My question is where I should put my defined macro variables as well as dataset. Thanks.

SeanZ · ‎12-04-2013

My bad. I should have looked it carefully. Yes. the problem is solved. Thanks.

SeanZ · ‎12-04-2013

thank you for your reply. I ran your code and get following error 199 ods html file='c:\temp\new1.html' style=sasweb; NOTE: Writing HTML Body file: c:\temp\new1.html ERROR: Physical file does not exist, c:\temp\new1.html. ERROR: No body file. HTML output will not be created. WARNING: No output destinations active. Thanks.

SeanZ · ‎12-04-2013

Hi Jagadishkatam, thank you for your answer. it seems that the variables I want to keep are not listed in one variable "name". There are bench of variables in my dataset and I only want to keep those variable with _x as the end. Your code to me seems to be one variable "name" including all variables' names. Do I have to first create one variable "name" to include all variable names? If so, how to create this variable? Thanks.

SeanZ · ‎11-22-2013

Hi, I have a dataset with hundreds of variables. I only want to keep some of them. Is there any easy way to pick them? Or I have to write their names separately. One example would be, I want to keep a1b1_x, a1b2_x,....a1b10_x, a2b1_x,.... These are 10X10 variables with _x at the end. How can I write it in the keep option? Thanks.

SeanZ · ‎11-19-2013

I used following code and wanted to take lag return value for each id each date, but the result is not desirable. Please check and let me know what's wrong. thanks data want; set have; by id date; if first.id then lagret = .; else lagret=lag(ret); run;

SeanZ · ‎11-15-2013

Hi Patrick. Thank you very much for your answer. That is enormous for me. I need some time to figure it out. But I do appreciate a lot for your work.

SeanZ · ‎11-15-2013

Hi jagadish. Yes, it will work if I close the session and open again. I don't want to do that since I have some dataset in the working directory and they are really big. I don't want to rerun them.

SeanZ · ‎11-15-2013

Do you mean run them and then re-run macro? I will definitely try it tomorrow. THanks.

SeanZ · ‎11-14-2013

I wrote a macro code. When I ran the code it returned some error. I corrected the error and tried to run it again. The problem is that it does not run. When I checked the log, it simply lists my code again, but not run it. What's the problem? Thanks.

SeanZ · ‎11-14-2013

I am not quite sure how to use data step to achieve this. The reason I want to use macro is because I need to roll the data (pick different sample for each loop). As I mentioned, for example, for the first loop I would like to grab one year's data from 1970/01/01 to 1970/12/31, after I get these year's data I can code and get one value, which is what I need. For the second loop, I would like to move forward to choose one year's data from 1970/02/01 to 1970/01/31 to calculate the second value. I would like to continue to move forward until sometime like 2010/1/1 to 2010/12/31 to get the last value. I want to create a variable and use macro to do a loop. something like %do i=&st %to &end, where st is starting point and end is the end point. I am not sure how to give an example since it's kind of complicated. But I hope you can understand what I want to achieve. I am not sure how to use data step. YOu can see that this will probably loop for more than 50 times.

SeanZ · ‎11-13-2013

i have a date variable from 1970/01/01 to 2010/12/31. And stock returns for each day. What I want to do is that use one year as a rolling window. Use full year's returns to calculate a new variable and merge that variable as the next month's date after one year window. Specifically, I will achieve two tasks. The first one, use returns from 1970/01/01 to 1970/12/31 to calculate one value and use this value as 1971/01's value(monthly), next, use returns from 1970/02/01 to 1970/01/31 to calculate the second Value and use this value as 1979/02's value, and so on. The second task would be: use returns from 1970/01/01 to 1970/12/31 to get one value as 1971/01/01's value(daily), next use returns from 1970/01/02 to 1971/01/01 to get the second value for 1970/01/02's value, and so on. Also, not all dates are available. Weekends and holidays are not in the data set. That means I only have around 250 days in each year. How to setup a macro to do this?

Online Status	Offline
Date Last Visited	‎10-11-2021 04:33 PM

how to choose a subsample based on list of values proc sql

Re: proc distance with missing data

Re: proc distance with missing data

Re: proc distance with missing data

Re: proc distance with missing data

Re: proc distance with missing data

proc distance with missing data

Calculate distance between two vectors

Re: backfill some observations

Re: backfill some observations

Re: proc distance with missing data

Re: proc distance with missing data

Re: Calculate distance between two vectors

Re: Calculate distance between two vectors

Re: how to define optional macro parameters?

Re: use proc sql to calculate some variables

Re: Full join in sql and keep the same joining variable

Full join in sql and keep the same joining variable

how to merge a dataset using macro loop

How to deal with macro variables with remote submission

Re: how to keep variables easier

Re: how to keep variables easier

Re: how to keep variables easier

how to keep variables easier

how to take lag value

Re: How to use a macro to do a rolling

Re: Macro does not run

Re: Macro does not run

Macro does not run

Re: How to use a macro to do a rolling

How to use a macro to do a rolling