I'm trying to follow this paper: Using a data science approach to predict cocaine use frequency from depressive symptoms where they use glm, gam with the beck inventory depression. So I did found a similiar dataset to test those models. However I'm having a hard time with both models. For example I have two variables d64a and d64b, and they're coded with 1,2,3,4 meaning that they're ordinal. Also, in the paper y2 is only the value of 1 but i have also a variable extra (that can be dependent, the proportion of consume)
For the GAM model I have:
b<-gam(y2~s(d64a)+s(d64b),data=DATOS2)
but I have the following error:
Error in smooth.construct.tp.smooth.spec(object, dk$data, dk$knots) :
A term has fewer unique covariate combinations than specified maximum degrees of freedom
Meanwhile for the glm, I have the following:
d<-glm(y2~d64a+d64b,data=DATOS2)
I don't know since d64a and d64b are ordinal I have to use factor()?
The error message tells you that one or both of d64a and d64b do not have 9 (nine) unique values.
By default s(...) will create a basis with nine functions. You get this error if there are fewer than nine unique values in the covariate.
Check which covariates are affected using:
length(unique(d64a))
length(unique(d64b))
and see what the number of unique values is for each of the covariates you wish to include. Then set the k argument to the number returned above if it is less than nine. FOr example, assume that the above checks returned 5 and 7 unique covariates, then you would indicate this by setting k as follows:
b <- gam(y2 ~ s(d64a, k = 5) + s(d64b, k = 7), data = DATOS2)
Related
For my class we have to create a model to predict the credit balance of each individuals. Based on observations, many results are zero where the lm tries to calculate them.
To overcome this I created a new variable that results in zero if X and Y are true.
CB$Balzero = ifelse(CB$Rating<=230 & CB$Income<90,0,1)
This resulted in getting 90% of the zero results right. The problem is:
How can I place this variable in the lm so it correctly results in zeros when the proposition is true and the calculation when it is false?
Something like: lm=Balzero*(Balance~.)
I think that
y ~ -1 + Balzero:Balance
might work (you haven't given us a reproducible example to try).
-1 tells R to omit the intercept
: specifies an interaction. If both variables are numeric, then A:B includes the product of A and B as a term in the model.
The second term could also be specified as I(Balzero*Balance) (I means "as is", i.e. interpret * in the usual numerical sense, not in its formula-construction context.)
These specifications should fit the model
Y = beta1*Balzero*Balance + eps
where eps is an error term.
If Balzero == 0, the predicted value will be zero. If Balzero==1 the predicted value will be beta1*Balance.
You might want to look into random forest models, which naturally incorporate the kind of qualitative splitting that you're doing by hand in your example.
I am trying to fit some data to a negative binomial model and run a pairwise comparison using emmeans. The data has two different sample sizes, 15 and 20 (num_sample in the example below).
I have set up two data frames: good.data which produces the expected result of offset() using random sample sizes between 15 and 20, and bad.data using a sample size of either 15 or 20, which seems to produce a factor of either 15 or 20. The bad.data pairwise comparison produces way too many comparisons compared to the good.data, even though they should produce the same number?
set.seed(1)
library(dplyr)
library(emmeans)
library(MASS)
# make data that works
data.frame(site=c(rep("A",24),
rep("B",24),
rep("C",24),
rep("D",24),
rep("E",24)),
trt_time=rep(rep(c(10,20,30),8),5),
pre_trt=rep(rep(c(rep("N",3),rep("Y",3)),4),5),
storage_time=rep(c(rep(0,6),rep(30,6),rep(60,6),rep(90,6)),5),
num_sample=sample(c(15,17,20),24*5,T),# more than 2 sample sizes...
bad=sample(c(1:7),24*5,T,c(0.6,0.1,0.1,0.05,0.05,0.05,0.05)))->good.data
# make data that doesn't work
data.frame(site=c(rep("A",24),
rep("B",24),
rep("C",24),
rep("D",24),
rep("E",24)),
trt_time=rep(rep(c(10,20,30),8),5),
pre_trt=rep(rep(c(rep("N",3),rep("Y",3)),4),5),
storage_time=rep(c(rep(0,6),rep(30,6),rep(60,6),rep(90,6)),5),
num_sample=sample(c(15,20),24*5,T),# only 2 sample sizes...
bad=sample(c(1:7),24*5,T,c(0.6,0.1,0.1,0.05,0.05,0.05,0.05)))->bad.data
# fit models
good.data%>%
mutate(trt_time=factor(trt_time),
pre_trt=factor(pre_trt),
storage_time=factor(storage_time))%>%
MASS::glm.nb(bad~trt_time:pre_trt:storage_time+offset(log(num_sample)),
data=.)->mod.good
bad.data%>%
mutate(trt_time=factor(trt_time),
pre_trt=factor(pre_trt),
storage_time=factor(storage_time))%>%
MASS::glm.nb(bad~trt_time:pre_trt:storage_time+offset(log(num_sample)),
data=.)->mod.bad
# pairwise comparison
emmeans::emmeans(mod.good,pairwise~trt_time:pre_trt:storage_time+offset(log(num_sample)))$contrasts%>%as.data.frame()
emmeans::emmeans(mod.bad,pairwise~trt_time:pre_trt:storage_time+offset(log(num_sample)))$contrasts%>%as.data.frame()
First , I think you should look up how to use emmeans.The intent is not to give a duplicate of the model formula, but rather to specify which factors you want the marginal means of.
However, that is not the issue here. What emmeans does first is to setup a reference grid that consists of all combinations of
the levels of each factor
the average of each numeric predictor; except if a
numeric predictor has just two different values, then
both its values are included.
It is that exception you have run against. Since num_samples has just 2 values of 15 and 20, both levels are kept separate rather than averaged. If you want them averaged, add cov.keep = 1 to the emmeans call. It has nothing to do with offsets you specify in emmeans-related functions; it has to do with the fact that num_samples is a predictor in your model.
The reason for the exception is that a lot of people specify models with indicator variables (e.g., female having values of 1 if true and 0 if false) in place of factors. We generally want those treated like factors rather than numeric predictors.
To be honest I'm not exactly sure what's going on with the expansion (276, the 'correct' number of contrasts, is choose(24,2), the 'incorrect' number of contrasts is 1128 = choose(48,2)), but I would say that you should probably be following the guidance in the "offsets" section of one of the emmeans vignettes where it says
If a model is fitted and its formula includes an offset() term, then by default, the offset is computed and included in the reference grid. ...
However, many users would like to ignore the offset for this kind of model, because then the estimates we obtain are rates per unit value of the (logged) offset. This may be accomplished by specifying an offset parameter in the call ...
The most natural choice for setting the offset is to 0 (i.e. make predictions etc. for a sample size of 1), but in this case I don't think it matters.
get_contr <- function(x) as_tibble(x$contrasts)
cfun <- function(m) {
emmeans::emmeans(m,
pairwise~trt_time:pre_trt:storage_time, offset=0) |>
get_contr()
}
nrow(cfun(mod.good)) ## 276
nrow(cfun(mod.bad)) ## 276
From a statistical point of view I question the wisdom of looking at 276 pairwise comparisons, but that's a different issue ...
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 a longitudinal dataset where I have the following variables for each subject:
IV: 3 factors (factorA, factorB, factorC, factorD), each measured twice, at the beginning and at the end of an intervention.
DV: one outcome variable (behavior), also measure twice, at the beginning and at the end of the intervention.
I would like to create a model that uses the change in factorA, factorB, factorC, factorD (change from beginning to end of the intervention) to predict the change in behavior (again from beginning to end).
I thought to use the delta values of factorA, factorB, factorC, factorD (from pre to post intervention) and use these delta values to predict the delta values of D1. I would also like to covary-out the absolute values of each factor (A, B, C and D) (e.g. using only the value at the beginning of the intervention for each factor) to make sure I account for the change that the absolute values (rather than the change) of these IVs may have on the DV.
Here is my dataset:
enter image description here
Here is my model so far:
Model <- lmer(Delta_behavior ~ Absolute_factorA + Absolute_factorB +
Absolute_factorC + Absolute_factorD + Delta_factorA +
Delta_factorB + Delta_factorC + Delta_factorD +
(1|Subject),a)
I think I am doing something wrong because I get this error:
Error: number of levels of each grouping factor must be < number of observations
What am I doing wrong? Is the data set structured weirdly? Should I not use the delta values? Should I use another test (not lmer)?
Because you have reduced your data to a single observation per subject, you don't need to use a multi-level/mixed model. The reason that lmer is giving you an error is that in this situation the between-subject variance is confounded with the residual variance.
You can probably go ahead and use a linear model (lm) for this analysis.
More technical detail
The equation for the distribution of the ith observation is something like [fixed-effect predictors] + eps(subject(i)) + eps(i) where eps(subject(i)) is the Normal error term of the subject associated with the ith observation, and eps(i) is the Normal residual error associated with the ith observation. If we only have one observation per subject, then each observation has two error terms that are unique to it. The sum of two Normal variables with zero means and variances of V1 and V2 is also Normal with mean zero and variance V1+V2 ... therefore V1 and V2 are jointly unidentifiable. You can use lmerControl to override the error if you really want to; lmer will return some arbitrary combination of V1, V2 estimates that sum to the total variance.
There's a similar example illustrated here.
What is the assign attribute of a linear model fit? It's supposed to somehow provide the position of the response term, but in practice it seems to enumerate all coefficients in the model. It's my understanding that assign is a carryover from S and it's not supported by glm(). I need to extract the equivalent information for glm, but I don't understand what the implementation does for lm and can't seem to find the source code either. The help file for lm.fit says, unhelpfully:
non-null fits will have components assign, effects and (unless not requested) qr relating to the linear fit, for use by extractor functions such as summary and effects
You find this in help("model.matrix"), which creates these values:
There is an attribute "assign", an integer vector with an entry for
each column in the matrix giving the term in the formula which gave
rise to the column. Value 0 corresponds to the intercept (if any), and
positive values to terms in the order given by the term.labels
attribute of the terms structure corresponding to object.
So, it maps the design matrix to the formula.
The numbers from $assign represent the corresponding predictor variable. If your predictor is categorical with 3 levels, you will see the corresponding number (3-1) times in your $assign call. Example:
data(mpg, package = "ggplot2")
m = lm(cty ~ hwy + class,data = mpg)
m$assign
[1] 0 1 2 2 2 2 2 2
# Note how there is six 2's to represent the indicator variables
# for the various 'class' levels. (class has 7 levels)
You will see the quantitative predictors will only have one value (hwy in the example above), since they are represented by one term in the design formula.