Dear R Community, I am plotting this simple x-y plot (raw data & plot attached). I cant fit a linear regression line to it. I have to figure out what is the best fit for this graph. Is there a way to tell which regression to use for this kind of data? Also, after selecting the best fit model, I need to extrapolate what could be the other possible data points. I am new to R. Could anyone please help? Thanks. Anuj http://www.nabble.com/file/p22336095/plot.jpg -- View this message in context: http://www.nabble.com/best-fit-line-tp22336095p22336095.html Sent from the R help mailing list archive at Nabble.com. [[alternative HTML version deleted]]
anujgoel wrote:> Dear R Community, > I am plotting this simple x-y plot (raw data & plot attached). > I cant fit a linear regression line to it. I have to figure out what is the > best fit for this graph. Is there a way to tell which regression to use for > this kind of data? > Also, after selecting the best fit model, I need to extrapolate what could > be the other possible data points. > I am new to R. Could anyone please help?Homework? PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Uwe Ligges> Thanks. > Anuj > http://www.nabble.com/file/p22336095/plot.jpg
On Mar 4, 2009, at 1:22 PM, anujgoel wrote:> > Dear R Community, > I am plotting this simple x-y plot (raw data & plot attached). > I cant fit a linear regression line to it. I have to figure out what > is the > best fit for this graph.That is virtually impossible to define rigorously. The "best fit" would go through all the points precisely, the extreme form of overfitting, but the mathematical result would not be informative at all. Where did these data come from? What domain of science are you working on? You want a result that incorporates the relationships known to exist in your domain of investigation and summarizes the data without overfitting. I am being intentionally vague in what follows because this looks like homework.> Is there a way to tell which regression to use for > this kind of data?Not really. Looks rather "hyperbolic", so you might think about the formula for hyperbola and then use the lm function.> > Also, after selecting the best fit model, I need to extrapolate what > could > be the other possible data points.The predict functions are used for this purpose. Consult you documentation. ?predict -- David Winsemius> > I am new to R. Could anyone please help? > Thanks. > Anuj > http://www.nabble.com/file/p22336095/plot.jpg > -- > View this message in context: http://www.nabble.com/best-fit-line-tp22336095p22336095.html > Sent from the R help mailing list archive at Nabble.com. > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.
Hi Anuj, Take a look at ?nls. It might be useful in this case. HTH, Jorge On Wed, Mar 4, 2009 at 1:22 PM, anujgoel <anujgoel@hotmail.co.uk> wrote:> > Dear R Community, > I am plotting this simple x-y plot (raw data & plot attached). > I cant fit a linear regression line to it. I have to figure out what is the > best fit for this graph. Is there a way to tell which regression to use for > this kind of data? > Also, after selecting the best fit model, I need to extrapolate what could > be the other possible data points. > I am new to R. Could anyone please help? > Thanks. > Anuj > http://www.nabble.com/file/p22336095/plot.jpg > -- > View this message in context: > http://www.nabble.com/best-fit-line-tp22336095p22336095.html > Sent from the R help mailing list archive at Nabble.com. > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >[[alternative HTML version deleted]]