An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20130710/3bad10b8/attachment.pl>
find a function for a random curve
3 messages · Xiaoyu Lu, S Ellison, Rolf Turner
I want to find a functional form for my data. I have tried smoothing and obtained a kinda perfect fit. However, I can only draw it but cannot call it.
You'll need to apply the smoothing function to the data separately, store that and plot the curve (?) fom that.
You can plot a line for many model fits using the associated predict() method and if necessary specifying a newdata data frame containing neatly-spaced predictors; look up the generic ?predict and predict.lm for a typical case. predict.nlm and predict.loess also exist and work much the same way.
S Ellison
*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}
Your question is very vague, so it's hard to know what to suggest. Please note the exhortation that appears at the bottom of every r-help message:
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
It is possible that splinefun() is what you are looking for.
cheers,
Rolf Turner
On 10/07/13 22:29, Xiaoyu Lu wrote:
Hi, I want to find a functional form for my data. I have tried smoothing and obtained a kinda perfect fit. However, I can only draw it but cannot call it. I am wondering that is there a way that I can store the fit as a function and apply it when needed. e.g. tranformation etc.