About MarkGIP

MarkGIP · ‎11-20-2011

Thank you so much, Rick! I'm really thinking about getting your book;)

MarkGIP · ‎11-18-2011

Hi! I know it's a little bit off topic but maybe you could help me..I spent some more time thinking about the correlation matrix. Do you know a way in IML to extract a correlation matrix from another (larger) correlation matrix? In detail, I want to select certain variables from a pool of variables. While the correlation matrix of the entire pool has been already calculated the correlation matrix of the selected variables should be extracted from the pool correlation matrix. Example: The pool consists of 10 variables. The corresponding 10x10 correlation matrix has been already calculated. Now I select three variables of the pool. How can I extract the 3x3 correlation matrix from the pool correlation matrix via IML? THANK YOU!!!

MarkGIP · ‎11-17-2011

Hi Rick! Once again I have to thank you for your quick response;) To be honest, I have't spent much time on thinking about the generation of a random correlation matrix yet. However, since my last post I've figured out how to use macros to loop the entire process. So the random correlation matrix is next on my list. Which of the structures you've mentioned would you prefer? So far I just thought about using uniformly distributed random numbers on the interval (-1,1), which I would use to create the correlation matrix. I'm thankful for every comment... THANK YOU!!!!

MarkGIP · ‎11-16-2011

Hi it's me again;) I would like to loop the RANDNORMAL function using a random correlation matrix for a certain number of times. The results for each drawing should be added horizontally to the already existing ones. Example: RANDNORMAL is used in combination with a random correlation matrix to create a 1000x100 matrix M of standard normal distributed variables with a specific correlation structure. The goal is to repeat the procedure 20 times. The results of each run should be merged together. The result would be a 1000x20000 matrix. Each 1000x100 block should have a specific correlation structure (determined by the correlation matrix which changes in every run). I hope you get the idea. I was wondering if you might know a way to loop the RANDNORMAL procedure in combination with a random correlation matrix.... THANKS A MILLION!!!

MarkGIP · ‎11-16-2011

Hi everyone! I'm quite new to SAS and I haven't worked with macros yet. Could you tell me if it's possible to loop an entire SAS program. With SAS program I mean different procedures such as DATA and IML. The program contains various random drawings and I would like to loop the program a certain number of times to get different results. The results for each run should be added to the previous results. Since I'm using different procedures I was wondering if there might be a "simple" macro to do the loop. Or is it smarter to do individual loops for every procedure? Thanks a million!!!

MarkGIP · ‎11-16-2011

Wow... that went fast! Thank you so much Rick!!!

MarkGIP · ‎11-16-2011

Hi everyone! Maybe you could help me with my latest obstacle I've encountered with IML. Do you happen to know how to convert a vector into a matrix? Here is a little example to give you a better idea: Given: Vector n (1000x1) Goal: Convert vector n into a 50x20 matrix m. The first column of matrix m should be equal to the first 50 elements of vector n. The second column should be equal to the next 50 elements and so on. It would be great if you had a solution for this problem... Thanks a million!!!

MarkGIP · ‎11-04-2011

Thank you Rick for your help. I will sure have a look at your blog;) Just one more thing... Is it also possible to replace the elements with elements of a different matrix (which has the same dimension like X)?

MarkGIP · ‎11-04-2011

Hi! I am fairly new to SAS and the concept of IML. Maybe you could help me with my problem: Is it possible to use the "if... then... else" function on the elements of the Matrix? My intention is to check whether the specific element of Matrix X(mxn) is greater than 2. If that is the case I'd like to change the value of the respective element in matrix X with 1. Else I want the value to be replaced by 0. This method should be applied to the entire Matrix. As a result I'd like to get a (mxn) matrix with just 0's and 2's I thought about using the "if" function: if x>2 then x=1; else x=0; However, this function won't allow me to check every element separately and keep the dimension of the original matrix. Looking forward to your support! THANK YOU!!!

MarkGIP · ‎11-04-2011

Works like a charm.... Thank you so much!!!!

MarkGIP · ‎11-04-2011

Sorry for confusing you... When I'm speaking of observations I mean records.... And yes the number of records is greater than two. Maybe there has been a slight mixup.... Please excuse if I'm a bit slow in the mind. But using your code on the entire records leads to an error. Do I have to peform the code on every record separately?

MarkGIP · ‎11-04-2011

Thank you all for your assistance! @Ksharp: Since I'm not familiar with macro I cannot check wether your code does the same... The problem with Art297's suggested code is that it does not seem to work when the variables have more then just one observation. Summing up the observations for each variable would not solve the problem cause I need them separated.

MarkGIP · ‎11-03-2011

@Art297 Aggregated data wouldn't help. To give you an idea: Let's assume my data consists of 5000 variables (a1-a5000) each with 100 records. My goal is to sum up those variables as follows (horizontal addition): x1= a1+...+a100 .....x50=a4900+...+a5000 Each new variable (x1 to x50) should have 100 records. Maybe this helps to clear things up! THANK YOU!

MarkGIP · ‎11-03-2011

Thank you all! @Rick: Thank you for the alternative method! @Art297: Wow.... Thank you so much! The comments really help a lot... But I still think that this code is not working in case of more than one record for each variable (see my last reply):(

MarkGIP · ‎11-03-2011

Thank you so much Steve!!! As it turns out, I forgot to state all variable names in the pctlpre option - I just put in letter q. So much for that.... Now there is only question 2 still to be solved....;)

Online Status	Offline
Date Last Visited	‎09-01-2015 07:11 AM

Re: Nesting Macros

Re: Nesting Macros

Re: Nesting Macros

Nesting Macros

Nesting Macros

Adding Rows and Columns to Matrix

Adding Rows and Columns to Matrix

Re: PROC SURVEYSELECT with Constraints

Adding Rows and Columns to Matrix

Re: PROC SURVEYSELECT with Constraints

Re: Quantiles and Sum of Variables...

Re: Loop RANDNORMAL

Re: Loop RANDNORMAL

Re: Loop RANDNORMAL

Loop RANDNORMAL

Loop an entire Program

Converting a Vector into a Matrix

Converting a Vector into a Matrix

Re: "If...then...else" Function for Matrix's Elements

"If...then...else" Function for Matrix's Elements

Re: Quantiles and Sum of Variables...

Re: Quantiles and Sum of Variables...

Re: Quantiles and Sum of Variables...

Re: Quantiles and Sum of Variables...

Re: Quantiles and Sum of Variables...

Re: Quantiles and Sum of Variables...