turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- How can we plot regression line to estimate the te...

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-01-2017 05:54 AM

Hi,

I am performing Noncompartmental Pharmacokinetics in SAS version 9.4 and for that this paper is really very useful.

http://www.lexjansen.com/pharmasug/2005/statisticspharmacokinetics/sp07.pdf

File required for NCA contains subjectID, Treatment, Time and concentration data.

Generally, terminal elimination constant is calculated from the linear regression of concentration vs. time curve using the least square regression method. For that minimum three points is neded to define the terminal (log-linear) portion of the curve and the selection is based on the best adjusted square coeffiecient of regression (r2).

How can we plot that regression line during the linear phase of the concentration vs. time data to estimate the terminal elimination rate constant (lambda z) using best fit method? How SAS automatically select number of required data points to calculate r2 and draw line to connect that specific points only ?

The desired output has been attached for reference.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-01-2017 11:04 AM

You will have to provide some guidelines for identifying the points you want. Maybe someone has a solution to something similar to this available but consider this purely fake data plotted below:

Which "linear" bit would you want? Time periods 1-3, 4-7 or 8-10?

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-06-2017 01:58 AM

Hello all,

In addition to mentioned query, I would like to elaborate it for better understanding. The point selection (i.e. it can be 3 or more then that) in graph is done automatically using best fit method. So, here the question is how it can be done in SAS? An image of graph has been attached for reference where based on number of data points selected automatically; R2, adjusted r2 and lambda z are calculated.