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
- /
- SAS Programming
- /
- General Programming
- /
- Excel LOGEST and LINEST estimation equation in SAS

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

11-26-2013 09:34 AM

Hi,

Anyone can help me how can I get the Excel's LOGEST and LINEST equation in sas.

Thanks,

KP

Accepted Solutions

Solution

11-26-2013
11:51 AM

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

11-26-2013 11:51 AM

Yes proc reg is what you need, and can be used for both.

You'll need to restructure your data though, so that you have 2 variables and 11 observations instead.

Here's an example of uing proc reg for exponential regression and there is an abundance of proc reg samples in the documentation and online.

http://www.sas.com/offices/NA/canada/downloads/presentations/Calgary/Linear_Regression.pdf

All Replies

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

11-26-2013 09:39 AM

What do those functions do?

PS. Please mark questions that are answered as Answered in the forums.

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

11-26-2013 11:12 AM

Thanks Reeza, I will do that.

It is LOGEST is Log Estimation and LINES is Linear Estimation - projecting the trend line.

I am somewhat sure that it is possible through Proc Reg, but not sure?

e.g.

if following data I am using to define the equation to estimate the future trend line:

Period | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 |

Coll | 88000 | 54000 | 65000 | 69000 | 55000 | 58000 | 59000 | 54000 | 53000 | 52000 | 51000 |

Logest will give me following equation: Coll = Alpha * (Beta ^ Period)

LOGEST | |

Beta | Alpha |

0.965071 | 73092.48 |

Linest will give me following equation: Coll = Alpha + Beta * (Period)

LINEST | |

Beta | Alpha |

-2318.18 | 73727.27 |

These equation I can use to forecast the trend.

Thanks,

KP

Solution

11-26-2013
11:51 AM

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

11-26-2013 11:51 AM

Yes proc reg is what you need, and can be used for both.

You'll need to restructure your data though, so that you have 2 variables and 11 observations instead.

Here's an example of uing proc reg for exponential regression and there is an abundance of proc reg samples in the documentation and online.

http://www.sas.com/offices/NA/canada/downloads/presentations/Calgary/Linear_Regression.pdf

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

11-27-2013 05:32 AM

Thanks Reeza