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

Showing results for

- JMP User Community
- :
- Discussions
- :
- JMP internal rate of return function

- 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
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 19, 2016 8:56 AM
(925 views)

1 ACCEPTED SOLUTION

Accepted Solutions

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 20, 2016 2:33 PM
(1655 views)

Solution

It appears that JMP requires a better "guess" than the Excel counterpart. It seems to work fine with a guess closer to the "true" IRR. For example, *5* instead of *0.1* works for most rows, but not all.

This formula seems to work for all rows in the sample table:

```
Internal Rate of Return(
(:real box office revenues + :real dvd earnings) / :real budget,
-:real budget,
:real box office revenues,
:real dvd earnings
);
```

7 REPLIES

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 19, 2016 1:29 PM
(916 views)

This is not an answer to your question but without the DVD in the formula the results look correct.

I thought it could be a formatting issue with the columns but changing them didn't solve the problem.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 19, 2016 2:17 PM
(908 views)

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 19, 2016 2:54 PM
(895 views)

I changed the second argument to a Sum Function, and it appeared to work.

Internal rate of return(.1,-:real budget,Sum(:real boxoffice revenues,::real dvd revenues))

Jim

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 19, 2016 4:09 PM
(882 views)

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 19, 2016 6:55 PM
(874 views)

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 20, 2016 2:33 PM
(1656 views)

It appears that JMP requires a better "guess" than the Excel counterpart. It seems to work fine with a guess closer to the "true" IRR. For example, *5* instead of *0.1* works for most rows, but not all.

This formula seems to work for all rows in the sample table:

```
Internal Rate of Return(
(:real box office revenues + :real dvd earnings) / :real budget,
-:real budget,
:real box office revenues,
:real dvd earnings
);
```

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Nov 21, 2016 6:51 AM
(808 views)

This works perfectly - so the formatting is not an issue, but only the initial guess. I would add that the guess is also an issue in Excel - the IRR calculation is wildly off if some estimated internal rates of return are negative and some are positive. To get around that in Excel, you need to make the initial guess either +0.1 or -0.1, depending on whether it will be a negative or positive (you can guess that by using the sum of the cash flows - whether these are positive or negative). It does appear that JMP is more finicky - the guess has to be closer to the right value, not just the right sign. Your use of the revenues/cost works quite well.

Thanks.