Paul, I think I would try something like yoyoras~te(temp,date) ... to incorportate a continuous interaction of temp and date. Otherwise with the models you are using, if `datecode' is a factor you'll get a smooth for every date, which may be over-flexible, and if datecode is numeric you'll get a linear dependance on date where the slope parameter changes with temp, which may be a bit too restrictive. Simon On Wednesday 17 June 2009 16:44, Paul Simonin wrote:> Hello R Users, > I have a question regarding fitting a model with GAM{mgcv}. I have > data from several predictor (X) variables I wish to use to develop a > model to predict one Y variable. I am working with ecological data, so > have data collected many times (about 20) over the course of two years. > Plotting data independently for each date there appears to be > relationships between Y (fish density) and at least several X variables > (temperature and light). However, the actual value of X variables (e.g., > temperature) changes with date/season. In other words, fish distribution > is likely related to temperature, but available temperatures change > through the season. Thus, when data from all dates are combined to > create a model from the entire dataset, I think I need to include some > type of metric/variable/interaction term to account for this date > relationship. I have written the following code using a "by" term: > > Distribution.s.temp.logwm2.deltaT<-gam(yoyras~s(temp,by=datecode)+s(logwm2, >by=datecode)+s(DeltaT,by=datecode),data=AllData) > > However, I am not convinced this is the correct way to account for > this relationship. What do you think? Is there another way to include > this in my model? Maybe I should simply include date ("datecode") as > another term in the model? > > I also believe there may be an interaction between temperature and > light (logwm2), and based on what I have read the "by" method may be the > best way to include this. Correct? > > Thank you for any input, tips, or advice you may be able to offer. I > am new to R, so especially grateful! > > Thanks again, > Paul Simonin > (PhD student) > > ______________________________________________ > 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.--> Simon Wood, Mathematical Sciences, University of Bath, Bath, BA2 7AY UK > +44 1225 386603 www.maths.bath.ac.uk/~sw283
Hello R Users,
I have a question regarding fitting a model with GAM{mgcv}. I have
data from several predictor (X) variables I wish to use to develop a
model to predict one Y variable. I am working with ecological data, so
have data collected many times (about 20) over the course of two years.
Plotting data independently for each date there appears to be
relationships between Y (fish density) and at least several X variables
(temperature and light). However, the actual value of X variables (e.g.,
temperature) changes with date/season. In other words, fish distribution
is likely related to temperature, but available temperatures change
through the season. Thus, when data from all dates are combined to
create a model from the entire dataset, I think I need to include some
type of metric/variable/interaction term to account for this date
relationship. I have written the following code using a "by" term:
Distribution.s.temp.logwm2.deltaT<-gam(yoyras~s(temp,by=datecode)+s(logwm2,by=datecode)+s(DeltaT,by=datecode),data=AllData)
However, I am not convinced this is the correct way to account for
this relationship. What do you think? Is there another way to include
this in my model? Maybe I should simply include date ("datecode") as
another term in the model?
I also believe there may be an interaction between temperature and
light (logwm2), and based on what I have read the "by" method may be
the
best way to include this. Correct?
Thank you for any input, tips, or advice you may be able to offer. I
am new to R, so especially grateful!
Thanks again,
Paul Simonin
(PhD student)
Hello R Users,
I have a question regarding fitting a model with GAM{mgcv}. I have data
from several predictor (X) variables I wish to use to develop a model to
predict one Y variable. I am working with ecological data, so have data
collected many times (about 20) over the course of two years. Plotting
data independently for each date there appears to be relationships
between Y (fish density) and at least several X variables (temperature
and light). However, the actual value of X variables (e.g., temperature)
changes with date/season. In other words, fish distribution is likely
related to temperature, but available temperatures change through the
season. Thus, when data from all dates are combined to create a model
from the entire dataset, I think I need to include some type of
metric/variable/interaction term to account for this date relationship.
I have written the following code using a "by" term:
Distribution.s.temp.logwm2.deltaT<-gam(yoyras~s(temp,by=datecode)+s(logwm2,by=datecode)+s(DeltaT,by=datecode),data=AllData)
However, I am not convinced this is the correct way to account for this
relationship. What do you think? Is there another way to include this in
my model? Maybe I should simply include date ("datecode") as another
term in the model?
I also believe there may be an interaction between temperature and
light (logwm2), and based on what I have read the "by" method may be
the
best way to include this. Correct?
Thank you for any input, tips, or advice you may be able to offer. I am
new to R, so especially grateful!
Thanks again,
Paul Simonin
(PhD student)