turn on suggestions

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

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- how to enter the initial value for parameters when fit a nonlinear model

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

Aug 16, 2013 1:33 PM
(1068 views)

Hello, I have a question about nonlinear model fit. For example, here is my data set is attached.

x y

0.861 | 1.000 |

0.775 | 1.082 |

0.686 | 1.007 |

0.896 | 0.985 |

0.796 | 1.007 |

0.690 | 1.000 |

0.917 | 0.971 |

0.807 | 0.977 |

0.722 | 0.970 |

0.714 | 0.898 |

0.587 | 0.774 |

0.456 | 0.564 |

0.448 | 0.384 |

0.321 | 0.191 |

0.201 | 0.115 |

I know it fits this model.

Y=1/(1+A*(EXP(B*X))

When I use JMP 10 to fit the model, I am stuck at the step what number should I enter to give an initial value for the parameter. I gave some numbers. It does run. Also, after I hit go, there is no plot. Can anyone help? Thanks a lot!

1 REPLY

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

Aug 19, 2013 7:02 AM
(997 views)

That's a logit there, so you can fit logit(Y) = a' + bx, and the values to insert for A will be exp(a') and for B, simply b.

Good luck.

Steve Denham