I am using the 'MuMIn' package in R to select models and calculate effect sizes of the input variables (rain, brk, onset, wid). To make my effect size comparable between variables, I standardised them using standardize function in arm package. Here is the code that I am following:
For reference, please refer to the appendix of this paper: http://onlinelibrary.wiley.com/doi/10.1111/j.1420-9101.2010.02210.x/full
Grueber et al. 2011: Multimodel inference in ecology and evolution: challenges and solutions
data1<-read.csv("data.csv",header=TRUE) #reads the data
global.model<-lmer(yld.res ~ rain + brk + onset + wid + (1|state),data=data1,REML="FALSE") # prepares a global model
stdz.model <- standardize(global.model,standardize.y = FALSE) # standardise the input varaibles
model.set <- dredge(stdz.model) ### generates the full submodel set
top.models <- get.models(model.set, subset= delta<2) # selects models with delta AIC <2
model.avg(top.models) # calculates the average effect size of input variables
Here is the result of model.avg(top.models) which gives the average effect size of each input variable
Coefficients:
(Intercept) brk rain wid onset
subset -4.281975e-14 -106.0919 51.54688 39.82837 35.68766
I read around how the standardize function works- subtracts mean and divides by 2SD.
My question is this: Since I have standardised the input variables, should not the effect sizes be between -1 to 1? or the effect size which the output shows is correct?
Please advise
Thanks a lot
This is more of a statistical question than a programming question, but: you've only standardized the predictor variables, not the response variable (you specified standardize.y=FALSE); therefore, each of your coefficients represents the expected change of the response (in the response's units!) per 2 SD change in the predictor. If the range of the response is large (as it must be in your example), then there could be a very large change. For example, if I were analyzing the change in elephant weight measured in milligrams, I could expect very large changes in the response for reasonably small changes in the predictors (e.g. sex, age, food availability). You should probably use standardize.y=TRUE if you want truly nondimensional/unitless effect sizes. Even nondimensional effects aren't necessarily constrained to be between -1 and +1, but it would be surprising for them to be so large.
By the way, I think your standardize function comes from the arm package, not from MuMIn (library("sos"); findFn("standardize",sortby="Function)).
Related
How to run Latent Class Growth Modelling (LCGM) with a multinomial response variable in R (using the flexmix package)?
And how to stratify each class by a binary/categorical dependent variable?
The idea is to let gender shape the growth curve by cluster (cf. Mikolai and Lyons-Amos (2017, p. 194/3) where the stratification is done by education. They used Mplus)
I think I might have come close with the following syntax:
lcgm_formula <- as.formula(rel_stat~age + I(age^2) + gender + gender:age)
lcgm <- flexmix::stepFlexmix(.~ .| id,
data=d,
k=nr_of_classes, # would be 1:12 in real analysis
nrep=1, # would be 50 in real analysis to avoid local maxima
control = list(iter.max = 500, minprior = 0),
model = flexmix::FLXMRmultinom(lcgm_formula,varFix=T,fixed = ~0))
,which is close to what Wardenaar (2020,p. 10) suggests in his methodological paper for a continuous outcome:
stepFlexmix(.~ .|ID, k = 1:4,nrep = 50, model = FLXMRglmfix(y~ time, varFix=TRUE), data = mydata, control = list(iter.max = 500, minprior = 0))
The only difference is that the FLXMRmultinom probably does not support varFix and fixed parameters, altough adding them do produce different results. The binomial equivalent for FLXMRmultinom in flexmix might be FLXMRglm (with family="binomial") as opposed FLXMRglmfix so I suspect that the restrictions of the LCGM (eg. fixed slope & intercept per class) are not specified they way it should.
The results are otherwise sensible, but model fails to put men and women with similar trajectories in the same classes (below are the fitted probabilities for each relationship status in each class by gender):
We should have the following matches by cluster and gender...
1<->1
2<->2
3<->3
...but instead we have
1<->3
2<->1
3<->2
That is, if for example men in class one and women in class three would be forced in the same group, the created group would be more similar than the current first row of the plot grid.
Here is the full MVE to reproduce the code.
Got similar results with another dataset with diffent number of classes and up to 50 iterations/class. Have tried two alternative ways to predict the probabilities, with identical results. I conclude that the problem is most likely in the model specification (stepflexmix(...,model=FLXMRmultinom(...) or this is some sort of label switch issue.
If the model would be specified correctly and the issue is that similar trajectories for men/women end up in different classes, is there a way to fix that? By for example restricting the parameters?
Any assistance will be highly appreciated.
This seems to be a an identifiability issue apparently common in mixture modelling. In other words the labels are switched so that while there might not be a problem with the modelling as such, men and women end up in different groups and that will have to be dealt with one way or another
In the the new linked code, I have swapped the order manually and calculated the predictions with by hand.
Will be happy to hear, should someone has an alternative approach to deal with the label swithcing issue (like restricting parameters or switching labels algorithmically). Also curious if the model could/should be specified in some other way.
A few remarks:
I believe that this is indeed performing a LCGM as we do not specify random effects for the slopes or intercepts. Therefore I assume that intercepts and slopes are fixed within classes for both sexes. That would mean that the model performs LCGM as intended. By the same token, it seems that running GMM with random intercept, slope or both is not possible.
Since we are calculating the predictions by hand, we need to be able to separate parameters between the sexes. Therefore I also added an interaction term gender x age^2. The calculations seems to slow down somewhat, but the estimates are similar to the original. It also makes conceptually sense to include the interaction for age^2 if we have it for age already.
varFix=T,fixed = ~0 seem to be reduntant: specifying them do not change anything. The subsampling procedure (of my real data) was unaffected by the set.seed() command for some reason.
The new model specification becomes:
lcgm_formula <- as.formula(rel_stat~ age + I(age^2) +gender + age:gender + I(age^2):gender)
lcgm <- flexmix::flexmix(.~ .| id,
data=d,
k=nr_of_classes, # would be 1:12 in real analysis
#nrep=1, # would be 50 in real analysis to avoid local maxima (and we would use the stepFlexmix function instead)
control = list(iter.max = 500, minprior = 0),
model = flexmix::FLXMRmultinom(lcgm_formula))
And the plots:
I am using the useful gratia package by Gavin Simpson to extract the difference in two smooths for two different levels of a factor variable. The smooths are generated by the wonderful mgcv package. For example
library(mgcv)
library(gratia)
m1 <- gam(outcome ~ s(dep_var, by = fact_var) + fact_var, data = my.data)
diff1 <- difference_smooths(m1, smooth = "s(dep_var)")
draw(diff1)
This give me a graph of the difference between the two smooths for each level of the "by" variable in the gam() call. The graph has a shaded 95% credible interval (CI) for the difference.
Statistical significance, or areas of statistical significance at the 0.05 level, is assessed by whether or where the y = 0 line crosses the CI, where the y axis represents the difference between the smooths.
Here is an example from Gavin's site where the "by" factor variable had 3 levels.
The differences are clearly statistically significant (at 0.05) over nearly all of the graphs.
Here is another example I have generated using a "by" variable with 2 levels.
The difference in my example is clearly not statistically significant anywhere.
In the mgcv package, an approximate p value is outputted for a smooth fit that tests the null hypothesis that the coefficients are all = 0, based on a chi square test.
My question is, can anyone suggest a way of calculating a p value that similarly assesses the difference between the two smooths instead of solely relying on graphical evidence?
The output from difference_smooths() is a data frame with differences between the smooth functions at 100 points in the range of the smoothed variable, the standard error for the difference and the upper and lower limits of the CI.
Here is a link to the release of gratia 0.4 that explains the difference_smooths() function
enter link description here
but gratia is now at version 0.6
enter link description here
Thanks in advance for taking the time to consider this.
Don
One way of getting a p value for the interaction between the by factor variables is to manipulate the difference_smooths() function by activating the ci_level option. Default is 0.95. The ci_level can be manipulated to find a level where the y = 0 is no longer within the CI bands. If for example this occurred when ci_level = my_level, the p value for testing the hypothesis that the difference is zero everywhere would be 1 - my_level.
This is not totally satisfactory. For example, it would take a little manual experimentation and it may be difficult to discern accurately when zero drops out of the CI. Although, a function could be written to search the accompanying data frame that is outputted with difference_smooths() as the ci_level is varied. This is not totally satisfactory either because the detection of a non-zero CI would be dependent on the 100 points chosen by difference_smooths() to assess the difference between the two curves. Then again, the standard errors are approximate for a GAM using mgcv, so that shouldn't be too much of a problem.
Here is a graph where the zero first drops out of the CI.
Zero dropped out at ci_level = 0.88 and was still in the interval at ci_level = 0.89. So an approxiamte p value would be 1 - 0.88 = 0.12.
Can anyone think of a better way?
Reply to Gavin Simpson's comments Feb 19
Thanks very much Gavin for taking the time to make your comments.
I am not sure if using the criterion, >= 0 (for negative diffs), is a good way to go. Because of the draws from the posterior, there is likely to be many diffs that meet this criterion. I am interpreting your criterion as sample the posterior distribution and count how many differences meet the criterion, calculate the percentage and that is the p value. Correct me if I have misunderstood. Using this approach, I consistently got p values at around 0.45 - 0.5 for different gam models, even when it was clear the difference in the smooths should be statistically significant, at least at p = 0.05, because the confidence band around the smooth did not contain zero at a number of points.
Instead, I was thinking perhaps it would be better to compare the means of the posterior distribution of each of the diffs. For example
# get coefficients for the by smooths
coeff.level1 <- coef(gam.model1)[31:38]
coeff.level0 <- coef(gam.model1)[23:30]
# these indices are specific to my multi-variable gam.model1
# in my case 8 coefficients per smooth
# get posterior coefficients variances for the by smooths' coefficients
vp_level1 <- gam.model1$Vp[31:38, 31:38]
vp_level0 <- gam.model1$Vp[23:30, 23:30]
#run the simulation to get the distribution of each
#difference coefficient using the joint variance
library(MASS)
no.draws = 1000
sim <- mvrnorm(n = no.draws, (coeff.level1 - coeff.level0),
(vp_level1 + vp_level0))
# sim is a no.draws X no. of coefficients (8 in my case) matrix
# put the results into a data.frame.
y.group <- data.frame(y = as.vector(sim),
group = c(rep(1,no.draws), rep(2,no.draws),
rep(3,no.draws), rep(4,no.draws),
rep(5,no.draws), rep(6,no.draws),
rep(7,no.draws), rep(8,no.draws)) )
# y has the differences sampled from their posterior distributions.
# group is just a grouping name for the 8 sets of differences,
# (one set for each difference in coefficients)
# compare means with a linear regression
lm.test <- lm(y ~ as.factor(group), data = y.group)
summary(lm.test)
# The p value for the F statistic tells you how
# compatible the data are with the null hypothesis that
# all the group means are equal to each other.
# Same F statistic and p value from
anova(lm.test)
One could argue that if all coefficients are not equal to each other then they all can't be equal to zero but that isn't what we want here.
The basis of the smooth tests of fit given by summary(mgcv::gam.model1)
is a joint test of all coefficients == 0. This would be from a type of likelihood ratio test where model fit with and without a term are compared.
I would appreciate some ideas how to do this with the difference between two smooths.
Now that I got this far, I had a rethink of your original suggestion of using the criterion, >= 0 (for negative diffs). I reinterpreted this as meaning for each simulated coefficient difference distribution (in my case 8), count when this occurs and make a table where each row (my case, 8) is for one of these distributions with two columns holding this count and (number of simulation draws minus count), Then on this table run a chi square test. When I did this, I got a very low p value when I believe I shouldn't have as 0 was well within the smooth difference CI across almost all the levels of the exposure. Maybe I am still misunderstanding your suggestion.
Follow up thought Feb 24
In a follow up thought, we could create a variable that represents the interaction between the by factor and continuous variable
library(dplyr)
my.dat <- my.dat %>% mutate(interact.var =
ifelse(factor.2levels == "yes", 1, 0)*cont.var)
Here I am assuming that factor.2levels has the levels ("no", "yes"), and "no" is the reference level. The ifelse function creates a dummy variable which is multiplied by the continuous variable to generate the interactive variable.
Then we place this interactive variable in the GAM and get the usual statistical test for fit, that is, testing all the coefficients == 0.
#GavinSimpson actually posted a method of how to get the difference between two smooths and assess its statistical significance here in 2017. Thanks to Matteo Fasiolo for pointing me in that direction.
In that approach, the by variable is converted to an ordered categorical variable which causes mgcv::gam to produce difference smooths in comparison to the reference level. Statistical significance for the difference smooths is then tested in the usual way with the summary command for the gam model.
However, and correct me if I have misunderstood, the ordered factor approach causes the smooth for the main effect to now be the smooth for the reference level of the ordered factor.
The approach I suggested, see the main post under the heading, Follow up thought Feb 24, where the interaction variable is created, gives an almost identical result for the p value for the difference smooth but does not change the smooth for the main effect. It also does not change the intercept and the linear term for the by categorical variable which also both changed with the ordered variable approach.
I have some high dimensional repeated measures data, and i am interested in fitting random forest model to investigate the suitability and predictive utility of such models. Specifically i am trying to implement the methods in the LongituRF package. The methods behind this package are detailed here :
Capitaine, L., et al. Random forests for high-dimensional longitudinal data. Stat Methods Med Res (2020) doi:10.1177/0962280220946080.
Conveniently the authors provide some useful data generating functions for testing. So we have
install.packages("LongituRF")
library(LongituRF)
Let's generate some data with DataLongGenerator() which takes as arguments n=sample size, p=number of predictors and G=number of predictors with temporal behavior.
my_data <- DataLongGenerator(n=50,p=6,G=6)
my_data is a list of what you'd expect Y (response vector),
X (matrix of fixed effects predictors), Z (matrix of random-effects predictors),
id (vector of sample identifier) and time (vector of time measurements). To fit random forest model simply
model <- REEMforest(X=my_data$X,Y=my_data$Y,Z=my_data$Z,time=my_data$time,
id=my_data$id,sto="BM",mtry=2)
takes about 50secs here so bear with me
so far so good. Now im clear about all the parameters here except for Z. What is Z when i go to fit this model on my actual data?
Looking at my_data$Z.
dim(my_data$Z)
[1] 471 2
head(my_data$Z)
[,1] [,2]
[1,] 1 1.1128914
[2,] 1 1.0349287
[3,] 1 0.7308948
[4,] 1 1.0976203
[5,] 1 1.3739856
[6,] 1 0.6840415
Each row of looks like an intercept term (i.e. 1) and values drawn from a uniform distribution runif().
The documentation of REEMforest() indicates that "Z [matrix]: A Nxq matrix containing the q predictor of the random effects." How is this matrix to be specified when using actual data?
My understanding is that traditionally Z is simply one-hot (binary) encoding of the group variables (e.g. as described here), so Z from the DataLongGenerator() should be nxG (471x6) sparse matrix no?
Clarity on how to specify the Z parameter with actual data would be appreciated.
EDIT
My specific example is as follows, i have a response variable (Y). Samples (identified with id) were randomly assigned to intervention (I, intervention or no intervention). A high dimensional set of features (X). Features and response were measured at two timepoints (Time, baseline and endpoint). I am interested in predicting Y, using X and I. I am also interested in extracting which features were most important to predicting Y (the same way Capitaine et al. did with HIV in their paper).
I will call REEMforest() as follows
REEMforest(X=cbind(X,I), Y=Y, time=Time, id=id)
What should i use for Z?
When the function DataLongGenerator() creates Z, it's a random uniform data in a matrix. The actual coding is
Z <- as.matrix(cbind(rep(1, length(f)), 2 * runif(length(f))))
Where f represents the length of the matrices that represent each of the elements. In your example, you used 6 groups of 50 participants with 6 fixed effects. That led to a length of 472.
From what I can gather, since this function is designed to simulate longitudinal data, this is a simulation of random effects on that data. If you were working with real data, I think it would be a lot easier to understand.
While this example doesn't use RE-EM forests, I thought it was pretty clear, because it uses tangible elements as an example. You can read about random effects in section 1.2.2 Fixed v. Random Effects. https://ademos.people.uic.edu/Chapter17.html#32_fixed_effects
Look at section 3.2 to see examples of random effects that you could intentionally model if you were working with real data.
Another example: You're running a cancer drug trial. You've collected patient demographics on a weekly basis: weight, temperature, and a CBC panel and different groups of drug administration: 1 unit per day, 2 units per day, and 3 units per day.
In traditional regression, you'd model these variables to determine how accurately the model identifies the outcome. The fixed effects are the explainable variance or R2. So if you've .86 or 86% then 14% is unexplained. It could be an interaction causing the noise, the unexplained variance between perfect and what the model determined was the outcome.
Let's say the patients with really low white blood cell counts and were overweight responded far better to the treatment. Or perhaps the patients with red hair responded better; that's not in your data. In terms of longitudinal data, let's say that the relationship (the interaction relationship) only appears after some measure of time passes.
You can try to model different relationships to evaluate the random interactions in the data. I think you'd be better off with one of the many ways to evaluate interactions systematically than a random attempt to identify random effects, though.
EDITED I started to write this in the comments with #JustGettinStarted, but it was too much.
Without the background - the easiest way to achieve this would be to run something like REEMtree::REEMtree(), setting the random effects argument to random = ~1 | time / id). After it runs, extract the random effects it's calculated. You can do it like this:
data2 <- data %>% mutate(oOrder = row_number()) %>% # identify original order of the data
arrange(time, id) %>%
mutate(zOrder = row_number()) # because the random effects will be in order by time then id
extRE <- data.frame(time = attributes(fit$RandomEffects[2][["id"]])[["row.names"]]) %>%
separate(col = time,
into = c("time", "id"),
sep = "\\/") %>%
mutate(Z = fit$RandomEffects[[2]] %>% unlist(),
id = as.integer(id),
time = time)) # set data type to match dataset for time
data2 <- data2 %>% left_join(extRE) %>% arrange(oOrder) # return to original order
Z = cbind(rep(1, times = nrows(data2)), data2$Z)
Alternatively, I suggest that you start with the random generation of random effects. The random-effects you start with are just a jumping-off point. The random effects at the end will be different.
No matter how many ways I tried to use LongituRF::REEMforest() with real data, I ran into errors. I had an uninvertible matrix failure every time.
I noticed that the data generated by DataLongGenerator() comes in order by id, then time. I tried to order the data (and Z) that way, but it didn't help. When I extracted all the functionality out of the package LongituRF, I used the MERF (multiple-effects random forest) function with no problems. Even in the research paper, that method was solid. Just thought it was worth mentioning.
I am trying to figure out the best way to display a list of 30+ coefficients on a regression of a continuous variable.
(This may belong more in CrossValidated, I am not sure.)
Here is my example:
library("nycflights13")
library(dplyr)
flights <- nycflights13::flights
flights<- sample_n (flights, 3000)
m1<- glm(formula = arr_delay ~ . , data = flights)
summary(m1)
An option is dwplot from dotwhisker
library(dotwhisker)
dwplot(m1)
As #BenBolker commented, by default, the dwplot scales regression coeffficients by 2 standard deviations of the predictor variable
Or if we need a data.frame/tibble, then use tidy from broom
library(broom)
tidy(m1)
This may help. You could select a specific coefficient with the following :
str(flights) # to print list of data features
coef(m1)["age"] # here I just suppose that you have an axis called "age", you could select as many features coefficients as you want. For this you coud use a vector of relevant axis.
You could have a look at :
extract coefficients from glm in R
tl;dr dwplot is still (a) right answer, but there's a lot to say about the details of how you're fitting this model (and why it takes a really really long time).
glm vs lm
You're using glm() to fit a linear model, which isn't incorrect (and which would allow you to generalize to problems with count or binary responses). However, it's overkill in this case — lm() will work just fine, and be faster [considerably faster when it comes to generating confidence intervals etc.]
system.time(m1 <- glm(formula = arr_delay ~ . , data = flights)) ## 6 seconds
system.time(m2 <- lm(formula = arr_delay ~ . , data = flights, x=TRUE)) ## 13 seconds
(the reason for including x=TRUE will be discussed below)
The time difference becomes more stark when tidying/computing confidence intervals:
setTimeLimit(elapsed=600)
system.time(tidy(m1, conf.int=TRUE)) ## gave up after 10 minutes
system.time(tt <- tidy(m2, conf.int=TRUE)) ## 3.2 seconds
Tidying glms by default uses MASS::confint.glm() to compute confidence intervals by likelihood profiling, which is more accurate than Wald (mean +/- 1.96*SE) intervals for non-Gaussian responses), but way slower.
modeling choices
One of the reasons that everything is so slow is that there are lots of parameters (length(coef(m2)) is 1761). Why?
Although there are only 19 columns in the input data frame (so we might naively expect 18 coefficients), 4 of them are categorical, so get expanded to indicator variables:
catvars <- names(flights)[sapply(flights,is.character)]
sapply(catvars, function(x) length(unique(flights[[x]])))
## carrier tailnum origin dest
## 15 1653 3 94
So, most of the coefficients come from modeling the departures of individual planes (tailnum) [table(table(flights$tailnum)) shows that in this subsample of the data, more than half of the planes are recorded only once ...] It might not make sense to include this variable (if I were going to use tailnum, I would treat it as a random effect, although that would add a lot of modeling complexity).
Let's proceed without tailnum (we will still have plenty of coefficients to worry about).
plotting
At this point we're doing approximately what dotwhisker::dwplot does, but doing it by hand for more flexibility (in particular, ordering the terms by value).
The next step (1) extracts coefficients/conf int etc.; (2) scales non-binary variables by 2SD (using an internal function from dotwhisker); (3) drops the intercept; (4) makes term a factor ordered by the coefficient value and computes whether the term is significant (i.e., whether the lower and upper CI limits are both above or both below zero).
tt <- (tidy(m3, conf.int=TRUE)
%>% dotwhisker::by_2sd(flights)
%>% filter(term!="(Intercept)")
%>% mutate(term=reorder(factor(term),estimate),
sig=(conf.low*conf.high)>1)
)
Plot:
(ggplot(tt, aes(x=estimate,y=term,xmin=conf.low,xmax=conf.high))
+ geom_pointrange(aes(colour=sig))
+ geom_vline(xintercept=0,lty=2)
+ scale_colour_manual(values=c("black","red"))
)
I am using multivariate GAM models to learn more about fog trends in multiple regions. Fog is determined by visibility going below a certain threshold (< 400 meters). Our GAM model is used to determine the response of visibility to a range of meteorological variables.
However, my challenge right now is that I'd really like the y-axis to be the actual visibility observations rather than the centered smoothed. It is interesting to see how visibility is impacted by the covariates relative to the mean visibility in that location, but it's difficult to compare this for multiple locations where the mean visibility is different (and thus the 0 point in which visibility is enhanced or diminished has little comparable meaning).
In order to compare the results of multiple locations, I'm trying to make the y-axis actual visibility observations, and then I'll put a line at the visibility threshold we're interested in looking at (400 m)
to evaluate what the predictor variables values are like below that threshold (eg what temperatures are associated with visibility below 400 m).
I'm still a beginner when it comes to GAMs and R in general, but I've figured out a few helpful pieces so far.
Helpful things so far:
Attempt 1. how to extract gam fit for each variable in model
Extracting data used to make a smooth plot in mgcv
Attempt 2. how to use predict function to reconstruct a univariable model
http://zevross.com/blog/2014/09/15/recreate-the-gam-partial-regression-smooth-plots-from-r-package-mgcv-with-a-little-style/
Attempt 3. how to get some semblance of a y-axis that looks like visibility observations using "fitted" -- though I don't think this is
the correct approach since I'm not taking the intercept into account
http://gsp.humboldt.edu/OLM/R/05_03_GAM.html
simulated data
install.packages("mgcv") #for gam package
require(mgcv)
install.packages("pspline")
require(pspline)
#simulated GAM data for example
dataSet <- gamSim(eg=1,n=400,dist="normal",scale=2)
visibility <- dataSet[[1]]
temperature <- dataSet[[2]]
dewpoint <- dataSet[[3]]
windspeed <- dataSet[[4]]
#Univariable GAM model
gamobj <- gam(visibility ~ s(dewpoint))
plot(gamobj, scale=0, page=1, shade = TRUE, all.terms=TRUE, cex.axis=1.5, cex.lab=1.5, main="Univariable Model: Dew Point")
summary(gamobj)
AIC(gamobj)
abline(h=0)
Univariable Model of Dew Point
https://imgur.com/1uzP34F
ATTEMPT 2 -- predict function with univariable model, but didn't change y-axis
#dummy var that spans length of original covariate
maxDP <-max(dewpoint)
minDP <-min(dewpoint)
DPtrial.seq <-seq(minDP,maxDP,length=3071)
DPtrial.seq <-data.frame(dewpoint=DPtrial.seq)
#predict only the DP term
preds <- predict(gamobj, type="terms", newdata=DPtrial.seq, se.fit=TRUE)
#determine confidence intervals
DPplot <-DPtrial.seq$dewpoint
fit <-preds$fit
fit.up95 <-fit-1.96*preds$se.fit
fit.low95 <-fit+1.96*preds$se.fit
#plot
plot(DPplot, fit, lwd=3,
main="Reconstructed Dew Point Covariate Plot")
#plot confident intervals
polygon(c(DPplot, rev(DPplot)),
c(fit.low95,rev(fit.up95)), col="grey",
border=NA)
lines(DPplot, fit, lwd=2)
rug(dewpoint)
Reconstructed Dew Point Covariate Plot
https://imgur.com/VS8QEcp
ATTEMPT 3 -- changed y-axis using "fitted" but without taking intercept into account
plot(dewpoint,fitted(gamobj), main="Fitted Response of Y (Visibility) Plotted Against Dew Point")
abline(h=mean(visibility))
rug(dewpoint)
Fitted Response of Y Plotted Against Dew Point https://imgur.com/RO0q6Vw
Ultimately, I want a horizontal line where I can investigate the predictor variable relative to 400 meters, rather than just the mean of the response variable. This way, it will be comparable across multiple sites where the mean visibility is different. Most importantly, it needs to be for multiple covariates!
Gavin Simpson has explained the method in a couple of posts but unfortunately, I really don't understand how I would hold the mean of the other covariates constant as I use the predict function:
Changing the Y axis of default plot.gam graphs
Any deeper explanation into the method for doing this would be super helpful!!!
I'm not sure how helpful this will be as your Q is a little more open ended than we'd typically like on SO, but, here goes.
Firstly, I think it would help to think about modelling the response variable, which I assume is currently visibility. This is going to be a continuous variable, bounded at 0 (perhaps the data never reach zero?) which suggests modelling the data as conditionally distributed either
gamma (family = Gamma(link = 'log')) for visibility that never takes a value of zero.
Tweedie (family = tw()) for data that do have zeroes.
An alternative approach would be to model the occurrence of fog; if this is defined as an event <400m visibility then you could turn all your observations into 0/1 values for being a fog event or otherwise. Then you'd model the data as conditionally distributed Bernoulli, using family = binomial().
Having decided on a modelling approach, we need to model the response. This should be done using a multiple regression type of approach, with a GAM including multiple predictors. This way you get to estimate the effect of each potential predictor variable on the response while controlling for the effects of the other predictors. If you just do this using a single predictor at a time, say dewpoint, that variable could well "explain" variation in the data that might be due to another predictor, windspeed say, and you wouldn't know it.
Furthermore, there may well be interactions between predictors that you'll want to control for if they exist, which can only be done in
Then, to finally get to the crux of your problem, having fitted the multi-predictor model to "explain" visibility, you will need to predict from the model for sets of likely conditions. To look at how the visibility varies with dewpoint in a model where other predictor variables have effects, you need to fix the other variables at some reasonable values; one option is to set them to their mean (or modal value in the case of any factor predictor variables), or some other value indicative of typically values for that variable. You'll have to use your domain knowledge for this.
If you have interactions in the model, then you'll need to vary the two variables in the interaction, whilst holding all other variable fixed at some values.
Let's assume you don't have interactions and are interested in dewpoint but the model also includes windspeed. The mean windspeed for the values used to fit the model can be found from the cmX component of the fitted model. Of you could just calculate this from the observed windpseed values or set it to some known number you want to use. Denote the fitted by m, and the data frame with your data in it by df, then we can create new data to predict at over the range of dewpoint, whilst holding windspeed fixed.
mn.windspd <- m$cmX['windspeed']
## or
mn.windspd <- with(df, mean(windspeed))
## or set it some some value
mn.windspd <- 10 # say
Then you can do
preddata <- with(df,
expand.grid(dewpoint = seq(min(dewpoint),
max(dewpoint),
length = 300),
windspeed = mn.windspd))
Then you use this to predict from the fitted model:
pred <- predict(m, newdata = preddata, type = "link", se.fit = TRUE)
pred <- as.data.frame(pred)
Now we want to put these predictions back on to the response scale, and we want a confidence interval so we have to create that first before back transforming:
ilink <- family(m)$linkinv
pred <- transform(pred,
Fitted = ilink(fit),
Upper = ilink(fit + (2 * se.fit)),
Lower = ilink(fit - (2 * se.fit)),
dewpoint = preddata = dewpoint)
Now you can visualised the effect of dewpoint on the response whilst keeping windspeed fixed.
In your case, you will have to extend this to keeping temperature constant also, but that is done in the same way
mn.windspd <- m$cmX['windspeed']
mn.temp <- m$cmX['temperature']
preddata <- with(df,
expand.grid(dewpoint = seq(min(dewpoint),
max(dewpoint),
length = 300),
windspeed = mn.windspd,
temperature = mn.temp))
and then follow the steps above to do the prediction.
For one or two variables varying I have a function data_slice() in my gratia package which will do the above expand.grid() stuff for you so you don't have to specify the mean values of the other covariates:
preddata <- data_slice(m, 'dewpoint', n = 300)
technically this finds the value in the data closest to the median value (for the covariates not varying). If you want means, then do
fixdf <- data.frame(windspeed = mn.windspd, temperature = mn.temp)
preddata <- data_slice(m, 'dewpoint', data = fixdf, n = 300)
If you have an interaction, say between dewpoint and windspeed then you need to vary two variables. This is pretty easy again with expand.grid():
mn.temp <- m$cmX['temperature']
preddata <- with(df,
expand.grid(dewpoint = seq(min(dewpoint),
max(dewpoint),
length = 100),
windspeed = seq(min(windspeed),
max(windspeed),
length = 300),
temperature = mn.temp))
This will create a 100 x 100 grid of values of the covariates to predict at, whilst holding temperature constant.
For data_slice() you'd need to do:
fixdf <- data.frame(temperature = mn.temp)
preddata <- data_slice(m, 'dewpoint', 'windpseed',
data = fixdf, n = 300)
And extending this on to more covariates you want to vary, is also easy following this pattern with expand.grid(); I have yet to implement more than 2 variables varying in data_slice.