Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Software & Technology
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

Browse Open Actuarial Jobs

Life  Health  Casualty  Pension  Entry Level  All Jobs  Salaries


Reply
 
Thread Tools Display Modes
  #201  
Old 05-20-2015, 01:05 PM
Pujols4Prez's Avatar
Pujols4Prez Pujols4Prez is offline
Member
SOA AAA
 
Join Date: Jul 2008
Posts: 12,029
Default

Quote:
Originally Posted by BassFreq View Post
I think that cannot be done in proc univariate. To get years in the x-axis, you would have to say "histogram years;". Then the only way to get the cashflows in would be to use them as a weight or frequency, but then you would have to ignore all negative cash flows.

Is a histogram of cash flows really what you're looking for?
Probably not, but I am trying to work with the tools I have. I want to fit a curve to a set of cashflows (all positive) and display why I think that distribution is a good fit visually.
__________________
Hardwork is the crutch of the talent-less.

Quote:
Originally Posted by Statatak View Post
what's a JAS
Reply With Quote
  #202  
Old 05-20-2015, 02:52 PM
BassFreq's Avatar
BassFreq BassFreq is offline
Member
CAS
 
Join Date: Jun 2003
Location: Chicago
Studying for all eternity
Favorite beer: Duff
Posts: 1,684
Blog Entries: 2
Default

If you want to fit a curve (as opposed to a predefined distribution), then you could try proc reg, genmod, gam, or loess, and use transformations of the year (year, year^2, log(year), etc) as regressors to predict the cash flows.
__________________
If at first you don't succeed, you have one data point.
Res ipsa loquitur, sed quid in infernos dicet?
Reply With Quote
  #203  
Old 05-20-2015, 03:31 PM
Pujols4Prez's Avatar
Pujols4Prez Pujols4Prez is offline
Member
SOA AAA
 
Join Date: Jul 2008
Posts: 12,029
Default

Quote:
Originally Posted by BassFreq View Post
If you want to fit a curve (as opposed to a predefined distribution), then you could try proc reg, genmod, gam, or loess, and use transformations of the year (year, year^2, log(year), etc) as regressors to predict the cash flows.
Yeah, I wanted to use proc reg but we don't have the SAS/STAT package. I might need to request purchasing it, but that is a lot of red tape.
__________________
Hardwork is the crutch of the talent-less.

Quote:
Originally Posted by Statatak View Post
what's a JAS
Reply With Quote
  #204  
Old 05-20-2015, 04:22 PM
BassFreq's Avatar
BassFreq BassFreq is offline
Member
CAS
 
Join Date: Jun 2003
Location: Chicago
Studying for all eternity
Favorite beer: Duff
Posts: 1,684
Blog Entries: 2
Default

Quote:
Originally Posted by Pujols4Prez View Post
Yeah, I wanted to use proc reg but we don't have the SAS/STAT package. I might need to request purchasing it, but that is a lot of red tape.
Given those circumstances, you might be better off summarizing the data in SAS, and then performing the curve fitting in R or or Excel.
__________________
If at first you don't succeed, you have one data point.
Res ipsa loquitur, sed quid in infernos dicet?
Reply With Quote
  #205  
Old 06-06-2015, 11:26 PM
ScottKelly's Avatar
ScottKelly ScottKelly is offline
Member
SOA
 
Join Date: Jul 2012
Posts: 285
Default

fyi there is a university edition to SAS that you can download for free for non commercial use
Reply With Quote
  #206  
Old 06-22-2015, 04:22 PM
Actinium ad Infinitum's Avatar
Actinium ad Infinitum Actinium ad Infinitum is offline
SOA AAA
 
Join Date: Jun 2015
Location: US
Favorite beer: Coffee 24x7
Posts: 13
Default

Quote:
Originally Posted by T-roy View Post
I also have "The Little SAS Book" by Delwiche and Slaughter. It is nice to have this as a reference, although it is kinda vague and pretty basic.
"The Little SAS Book" is a good starter reference.
Reply With Quote
  #207  
Old 06-22-2015, 04:25 PM
Actinium ad Infinitum's Avatar
Actinium ad Infinitum Actinium ad Infinitum is offline
SOA AAA
 
Join Date: Jun 2015
Location: US
Favorite beer: Coffee 24x7
Posts: 13
Default Proc SQL

If you are doing a lot of Data Steps, you may want to consider using PROC SQL. In many cases, PROC SQL requires less coding and is more efficient.

Last edited by Actinium ad Infinitum; 06-22-2015 at 04:57 PM..
Reply With Quote
  #208  
Old 10-18-2015, 04:11 PM
ActSciMan's Avatar
ActSciMan ActSciMan is offline
Member
CAS
 
Join Date: Nov 2010
Studying for my family
Posts: 1,836
Default

Quote:
Originally Posted by Actinium ad Infinitum View Post
If you are doing a lot of Data Steps, you may want to consider using PROC SQL. In many cases, PROC SQL requires less coding and is more efficient.
And, alternatively, in many other cases, the data step requires less coding and is more efficient.

The book Proc SQL by Example is good. He teaches you PROC SQL, but with just about everything he does, he shows you how to do the same thing with the data step, or explains that you can't and how close you can get, or shows you things you can't do with PROC SQL but can with the data step.
Reply With Quote
  #209  
Old 08-11-2017, 06:22 PM
SanctuS's Avatar
SanctuS SanctuS is offline
Member
 
Join Date: Apr 2008
Posts: 63
Default

I don't know if anyone will see this, but I have a question.

I inherited a 2 part process in SAS Enterprise Guide. If I run a step in Part 1 I define a directory as something like

Quote:
%Let Butts = \\FILEPATH
Then in the second program in the process I import data into my work directory

Quote:
Proc Import
DATAFILE = "FILENAMES_ARE_FOR_NERDS"
OUT = work.NERD_STUFF
RUN;
Next step goes to look for that dataset, but I get errors:

Quote:
DATA NERDS_FOREVER
SET NERD_STUFF
x
RUN;
Get errors that say:
Quote:
BUTTS.NERD_STUFF.DATA does not exist.
Somehow it's assigning the work directory to BUTTS, when it's intended to be the Temp directory.

If I run Part 1. Close SAS. Open SAS. Run Part 2. I get by with no issues.

Does anyone have any idea why SAS would not know how to find it's own work/temp directory and would look for BUTTS?
Reply With Quote
  #210  
Old 08-11-2017, 06:32 PM
MoralHazard MoralHazard is offline
Member
CAS
 
Join Date: Jul 2011
Favorite beer: Sam Adams Rebel Rouser
Posts: 86
Default

Is there an "options user=butts" somewhere in your code?
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


All times are GMT -4. The time now is 02:49 AM.


Powered by vBulletin®
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
*PLEASE NOTE: Posts are not checked for accuracy, and do not
represent the views of the Actuarial Outpost or its sponsors.
Page generated in 0.17110 seconds with 10 queries