10
votes
Accepted
Zero inflated beta regression using gamlss for vegetation cover data
I have added preliminary support for gamlss to the emmeans package...
...
- 19.2k
10
votes
Accepted
Simulate linear regression with heteroscedasticity
To simulate data with a varying error variance, you need to specify the data generating process for the error variance. As has been pointed out in the comments, you did that when you generated your ...
7
votes
Accepted
Are Random Forests more powerful than generalized linear models?
You should try lots of models. The 'no free lunch' theorem states that there is no one best model - every situation is different. Logistic regression for example is desirable when it works because ...
- 1,762
7
votes
Are there better approaches than the weighted mean?
There seems to be a smooth dependence of variance on observation index, so you could try a joint modeling approach, see for instance Articles that work with covariates for mean, variance, and ...
- 74.5k
6
votes
Accepted
Is there a hypothesis test that tells us whether we should use GAM vs GLM?
Simon Wood, the author of the mgcv package for R and a statistician who has made significant contributions to GAM theory and methods, has developed well-performing Wald-like tests for smooths. As such,...
- 45.5k
5
votes
fit GLM for weibull family
Sorry i'm quite late with this.... but might help someone i believe :
gamlss package is what you should be looking for. It supports almost all the distributions( ...
- 243
5
votes
Are Random Forests more powerful than generalized linear models?
One point to consider is are you interested in making predictions or understanding associations and carrying out inference (confidence intervals around effects). Although random forests provide a ...
- 2,312
5
votes
Simulate linear regression with heteroscedasticity
You need to model the heteroskedasticity. One approach is via the R package (CRAN) dglm, dispersion generalized linear model. This is an extension of glm's which, ...
- 74.5k
5
votes
Accepted
Is smoothing an appropriate solution to deal with model diagnostics in a GAMLSS?
The overall and predictor-specific worm plots share the feature that "different shapes indicate different inadequacies in the model", as explained in the article Analysis of longitudinal ...
- 19.7k
5
votes
Accepted
How to compare centiles from different models?
Some of the deviations you point to can be due to random variation alone. There are simple ways to test them.
When there are independent observations of a variable whose conditional distributions ...
- 316k
5
votes
Accepted
Difference between ZIP and ZIP2
There's more details in the freely available older version of the gamlss book (section 18.2.11-12). The regular Poisson model is parametrized through a mean $\mu$, ...
- 1,826
4
votes
How can GAMLSS relax the GLM exponential family assumption?
What Rigby and Stasinoplous' GAMLSS models allow is the modelling of all parameters of a distribution with separate linear predictors. Thomas Yee's Vector Generalised Additive Model (VGAM) class of ...
- 45.5k
4
votes
What does it mean to perform regression using a specific distribution?
These concepts fall under the subject of Generalized Linear Models. Generalized linear models contain two components
A random component. The observed data is distributed according to some ...
- 35.2k
4
votes
Accepted
What does it mean to perform regression using a specific distribution?
Regression models aim to estimate a parameter (typically a mean) for a response variable conditional on a (set of) regressor variable(s). To do that, we generally need to specify the nature of the ...
3
votes
fit GLM for weibull family
The glm() function does not support the Weibull distribution in R unfortunately. You can try ?family to see which distributions ...
- 213
3
votes
Stepwise model selection using Generalized Akaike Information Criterion
In general, you can't select "the best" model using stepwise regression. All statistics produced through stepwise model building have a nested chain of invisible/unstated "conditional on excluding X" ...
- 29.2k
3
votes
Accepted
Create Spline from Coefficients and Knots in GAMLSS
the pb() function fits P-splines as described by Eilers and Marx (1996): B-splines on equally spaced knots and finite difference penalties. In the same paper there are some code chunks that show how ...
- 1,151
3
votes
Accepted
What distribution has exactly three parameters for mean, variance, and skewness?
Infinitely many answers are possible, some links in comments. One family is the skew-normal, but it only admits limited degrees of skewness. The same ideas used to construct the skew-normal family ...
- 74.5k
3
votes
Accepted
Is it possible to use location-scale family distributions for mixed effects modeling?
A couple of comments:
Generalized Linear Mixed Models (GLMMs) have the following general representation:
$$\left\{
\begin{array}{l}
Y_i \mid b_i \sim \mathcal F_\psi,\\\\
b_i \sim \mathcal N(0, D),
\...
- 19.9k
3
votes
Is smoothing an appropriate solution to deal with model diagnostics in a GAMLSS?
A worm plot is basically a qq plot, so what you are doing is trying to find the best functional form of the covariates that yields a normal quantile Residual. This indicates a better fit.
You checked ...
- 1,448
3
votes
Accepted
How to transform an uniform distribution into a generalized beta 2 distribution using gamlss, fitdist or other?
The generic way to solve this problem (converting a set of uniform random deviates to an alternative probability distribution) is to use the inverse cumulative distribution function or quantile ...
- 41.3k
3
votes
Accepted
Predict gamlss one-inflated beta model
The predicted probabilities that Y=1 are given by
p1 = nu/(1+nu)
So just predict nu and then transformation nu to p1.
- 318
3
votes
Accepted
can daily count data use GAM ordered categorical family, proportional-odds model?
For smoothing functions in gamlss I usually use
P-splines, e.g.
pb(Time),
where the smoothing parameter is estimated automatically
using a local maximum likelihood ...
- 318
3
votes
How to interpret the p-value associated with the intercept?
I want to establish which interactions of levels of factors are significantly different from the intercept.
You will get into trouble if you only depend on the coefficients and p values displayed ...
- 86.4k
2
votes
Modelling zero-inflated proportion data in R using GAMLSS
I think the odds $\frac{p_0}{1-p_0-p_1}$ are given by $e^{\nu}$ not $\nu$, and similarly for $\frac{p_1}{1-p_0-p_1} = e^{\tau}$!
This means that in the above answer one needs to use the exponentials ...
2
votes
Confidence intervals with gamlss package
It sounds like you want prediction intervals. The se.fit argument name reflects that it will give you standard errors extracted from the original fit, and it sounds ...
- 20.7k
2
votes
Accepted
How to choose correct distribution in R
The possibility of values being exactly 0 and 1 would seem to rule out the just using the beta for all cases (otherwise I'd have suggested it earlier).
One possibility is to use a zero-one inflated ...
- 277k
2
votes
How to model heteroscedasticity and get the coefficients?
First, you need to formally define the model you want to estimate. Based on your example, you'll want to set a model like this:
$$Y_i = \beta_0 +\beta_1 X_i + \epsilon_i$$
where $\epsilon_i \sim N(0, ...
- 1,267
2
votes
Are Random Forests more powerful than generalized linear models?
Let's have some simple examples to show the differences. Our example have a single independent variable x and a single dependent variable - either real ...
- 121
2
votes
Smoothed Moments as Function of Predictor
Continuing from the example data generated in the OP, we can construct a simple GAMLSS model for the mean of the data using a penalised B-spline. This model assumed a Normal distribution. We are only ...
- 554
Only top scored, non community-wiki answers of a minimum length are eligible