How can i implement decision tree in SAS

sas
decision_trees

#1

Hello,

I am trying to solve the Titanic Problem in Kaggle through SAS.I have done in it R and used the rpart package to impute missing values.However I could not find any relevant procedure in SAS to do the same.
Came across the DTREE procedure but not sure if that would work.
Can someone please help me with this?
What other method can I use in SAS to impute the missing value of Age??


#2

Which version of SAS are you using? The University edition of SAS does not allow creating decision tree.

You either need the Operations module or access to SAS Enterprise Miner.

Regards,
Kunal


#3

@data_hacks,

You can impute missing value of Age with mean of Age using Proc SQL and Data Step. Look at the commands below:

Identify Values to Impute Using General Case Method (Average of Age):

Imputation Using Data Step

Above, you have seen one of the methods to deal with it. You can also use multiple methods using SAS statements.

Hope this helps!

Regards,
Sunil


#4

Dear All,
For titanic dataset i am using SAS-Enterprise miner ,I am using the logic as follows as variable age has missing values. To predict missing values, we should use the salutation (Master, Mr, Miss, Mrs) of name as a new variable .

Can anybody guide about what modifications should i need to do with the IMPUTATION node ??