Questions tagged [generalized-linear-model]
A generalization of linear regression allowing for nonlinear relationships via a "link function" and for the variance of the response to depend on the predicted value. (Not to be confused with "general linear model" which extends the ordinary linear model to general covariance structure and multivariate response.)
4,577
questions
1
vote
1
answer
49
views
Predictive capacities of Generalized Linear Models and significance
I have the following data
...
0
votes
0
answers
23
views
Checking expected counts when analyzing binary outcome and binary explanatory variable using GLM
Understand that chi-squared test comparing two categorical variables is only valid when expected counts for each cell is at least 5. I have not read or heard that this needs to be checked when running ...
1
vote
1
answer
42
views
Mixed effect models where one fixed effect leads to very different outcomes [closed]
I am running a pilot experiment that is testing whether a modified form of music notation results in fewer errors in performance than conventional music notation.
Our experiment involves participants ...
0
votes
0
answers
20
views
Incomplete formula for the glm negative binomial
My data is about the number of insects that go to treatment 1 or treatment 2, and I have the factor place, with four levels (places). Each place has a treatment and ...
3
votes
1
answer
114
views
What is fundamentally different between a logistic regression and a Logit Discrete Choice model?
I'm trying to understand the difference between the two. I know there is a background random utility theory in discrete choice modeling, but I can't pinpoint, in what I've read, the difference between ...
0
votes
1
answer
29
views
How to calculate probability from logistic regression using interaction effect parameter?
I have a logistic regression (logit link) model that I use to estimate probabilities. In the absence of interaction effects this formula is given in a lot of references as
$$p_i = \frac{1}{1 + e^{-\...
0
votes
0
answers
36
views
Warning: Hauck-Donner effect detected in the following estimate(s): INTERCEPT
This is a quick question because I could not find an answer online.
I ran a truncated Poisson model and I got the warning in the title (perfect separation?) affecting my intercept.
The output is:
<...
1
vote
0
answers
18
views
Assumptions and validity of post-hoc pairwise comparisons of GLMs
How does packages for pairwise comparisons such as multcomp's glht(mcp) and emmeans pairs() maintain validity in cases of ...
2
votes
1
answer
48
views
how to interpret a generalised linear mixed model with binomial data
I know there has been a similar question before but I'm struggling to use the answers there to help interpret my data. I'm new to statistics so am very keen for help! I have data where each row ...
2
votes
0
answers
34
views
Comparing results from Poisson regressions
I have a dataset with count data (y variable), one primary variable of interest (factor denoted x_1), and several variables I wish to control for (factor and numeric denoted x_i).
Say, y is the number ...
0
votes
0
answers
29
views
How to interpret the coefficients of a logistic regression on a proportion?
Further to my previous post , it seems that one can/should use a logistic regression to model a proportion.
How do I interpret the coefficients of a logistic regression when the outcome variable is a ...
0
votes
0
answers
21
views
How to report predictors excluded from a GLM in a paper?
I am new to generalised linear models and want to use them for my fourth year dissertation project in ecology. Please forgive any ignorance on my part, I have done my best to research this on my own ...
0
votes
0
answers
10
views
Is it possible to reproduce a nested logit discrete choice model with a glm?
I'm trying to understand if I can fit mlogit's nested logit model using a different parametrization under the base glm procedure....
1
vote
0
answers
56
views
How to run fixed-effects model on survey data in R?
I have two periods of panel data and I am trying to see if coefficients change over time. I would like to control for individual-level heterogeneity, but not sure how to do it with a complex design ...
5
votes
1
answer
113
views
Percentage of females as a regressor in count model with population as offset - interpretation?
I am using a hurdle model (binomial logit for the zero counts, truncated Poisson for the positive counts) to study the effects of demographic (and other) variables on the adoption of private air-...
0
votes
0
answers
7
views
Handling Informative Censored Data in Comparative Analysis of Cumulative Outcomes Across Groups
Hello Cross Validated community,
I am working on a dataset that involves multiple groups undergoing a series of interventions, with the aim to achieve a binary outcome (e.g., success or failure). The ...
2
votes
1
answer
67
views
transforming a continous respond variable into a discrete one
I am visually estimating fish length underwater to convert it in fish biomass by means of proper length-weight relationships. Obviously, my accuracy is not perfect. When making hypothesis about ...
1
vote
0
answers
35
views
Interpretation of summary of a hierarchical logistic regression
I'd like to know if my interpretation of this summary is correct.
It is a hierarchical logistic regression I made with mixed models for an analysis of a generalization gradient. I've got fixed and ...
3
votes
1
answer
150
views
Understand and specify a generalized logistic model in R
I am reading a paper in which the authors models tree survival (mortality). They go and remeasure tagged trees for decades to establish "survival functions" for the given tree species and ...
0
votes
0
answers
8
views
Best method for building models using cumulative autocorrelated data
context:
As an example, say I am trying to find the causes of maintenance on a vehicle using its use. With an ultimate aim of understanding the relationship between the two.
If I only have variables ...
2
votes
1
answer
92
views
How does non-collapsibility and the lack of an error term affect coefficients in regression
I have read from here that in nonlinear models such as the logit and Cox, because of a lack of an error term, coefficients may be biased (typically towards zero) when covariates are omitted; I see how ...
4
votes
1
answer
172
views
Modelling a proportion
My outcome variable, 'sensitivity', is a continuous proportion ranging from 0 to 1, inclusive. For example, it indicates the percentage of instances in which my gold detector correctly identified the ...
1
vote
2
answers
67
views
Orthogonal or Raw Polynomial regression?
I am carrying out an analysis where I want to test for a relationship between two biological variables. The response is proportional (so I'm using a binomial GLM with weights) and the explanatory ...
2
votes
1
answer
51
views
Goodness of fit in GLM produces NANs
I have a GLM model which has a quadratic term and looks like this.
...
1
vote
0
answers
33
views
How to test the effects of multiple factors on classification accuracy? (Incomplete, unbalanced data set.)
I have a data set of classification results that are binary (correct prediction or false prediction).
Each entry stems from one of multiple models and contains a few more factors:
...
2
votes
0
answers
63
views
Which dependent variable is mostly impacted by predictor?
Usually one wants to identify the most important predictors (x1, x2, x3..., xn) in a regression model. My question is reversed: I have a data set that contains a risk factor risk and several outcomes ...
1
vote
1
answer
79
views
In a GLM, how do the dimensions of the linear predictor and the range of the link function always align?
Let $\mathbf{\vec y}$ be the response vector.
Then, we can write the exponential family as :
$$
\large p(y;\boldsymbol{\eta})=h(y) \exp \left(\boldsymbol{\eta} \cdot \mathbf{T}(y)-A(\boldsymbol{\eta})\...
2
votes
1
answer
29
views
When to use parametric/GLM/ANOVA, in relation to data normality
In my experiment we looked at colour change (luminance) following a predatory stimulus (presence and lock of being the two treatments). The data consists of luminance before, after and the % change ...
6
votes
1
answer
102
views
Is there a model for both mean and variance?
Currently we have models where $y^{pred}_{i} \sim N(\beta_1 x_i + \beta_0, \sigma^2)$.
Is it possible to create a model with non-constant variance $y^{pred}_{i} \sim N(\beta_1 x_i + \beta_0, e^{\...
0
votes
1
answer
54
views
Statistical Genetics - How to interpret data & analysis
My question is how to interpret $G_{ij}$ in the following statistical genetic model $$h(\mu_i) = \alpha_0 + \alpha'X_i + \beta'G_i,$$
where $G_i = (G_{i1},\dots ,G_{im})$ are allele counts (zero, one, ...
0
votes
0
answers
63
views
How to predict average Y from averages of X for GLM Quasi-poisson regression?
Need some help from the community :)
Data and Task Description
I have the data in following form:
Y
X1
X2
0
1000
10
2
1500
200
1
1000
110
Y - is the target variable I'm trying to predict. That is ...
4
votes
1
answer
102
views
R: Parameterization differences betwen MASS::glm.nb and glmmTMB "nbinom2"
I'd like opinions on two differing GLM outputs in RStudio.
I model count data (dung pellets) over 21 sites, using quadrats counted as an area offset. I started with a GLM Poisson regression for the ...
2
votes
0
answers
90
views
Why not always use covariate instead of offset in Poisson Regression? [duplicate]
I've just started studying Poisson regression and came across the two models:
$$
\begin{align*}
\log{\mathbb{E}(count)} &= \beta_0 + \beta_1x_1 + \beta_2x_2 + \log(T) \\\\
\log{\mathbb{E}(count)} &...
1
vote
0
answers
60
views
Linear model for maximizing rank correlation between observed and predicted response
linear regression is modelled as
$$Y = X\beta + \epsilon$$
for response variable $Y$ (vector), design matrix $X$, and iid Gaussian noise $\epsilon$ (vector).
instead of minimizing the mean squared ...
1
vote
1
answer
35
views
Availability of Linear Grouping Algorithms to Linearly Cluster Datasets
I have been trying to cluster a scatter plot that has a triangular graph, ideally the proper clustering plot should have a linear form, as shown below:
I tried using Spectral Clustering:
and ...
0
votes
1
answer
40
views
How can I calculate residuals of a dependent binary variable, using a glm (logistic) model that was fit on a different sample?
I have a data frame D1 in R with a dependent binary variable Response (0/1) and a set of covariates like age and gender. I want to know how "typical" ...
1
vote
0
answers
59
views
How to mathematically prove the "transitive property of nested predictors"?
QUESTION
I am studying the structure of experiment data sets, and I want to propose a rule that I call the "transitive property of nested predictors".
The general idea is that…
if there are ...
1
vote
1
answer
45
views
GLMM on proportion data based on counts
I am running a GLMM on my small dataset ($n=31$) in a repeated measures study that has $2$ groups and $5$ conditions (conditions are fixed for everyone). I am interested in main effects of group and ...
2
votes
1
answer
81
views
0
votes
1
answer
106
views
Impact of outliers to QQ plot
I'm trying to build an GLM regression (10k samples and 50 dimensions). I ran an analysis of the dependent variable since the regression has a normality assumption for the dependent variable.
The QQ ...
5
votes
1
answer
68
views
difference between GLM covariance matrix from MLE vs. IRLS for non-canonical link
Someone asked a question on Stack Overflow where they noted a difference between Minitab and R (glm) results for the variance-covariance matrix of the parameters, ...
0
votes
0
answers
18
views
correct set up for fitting binomial data in GLM
I have the following data and am trying to fit a glm model. There are six factors. Each factor has $250$ samples and $Y$ accounts for the successes.
...
3
votes
2
answers
93
views
lmer analysis for 58 people 4 timepoints, advice on setting the model
I’m looking for some advice on analysing this data. I think I need a multilevel model.
I have 58 participants measured progesterone and a factor at 4 timepoints. I would like to do a model to see how ...
0
votes
0
answers
21
views
Does rolling average help reduce generalize linear model on timeseries variable to OLS?
I have a data $Y_t$ that basically measures the number of certain events at any day, and I have a time-series for all countries from 2015 to now. And I am trying to fit that $Y_t$ to some $X_t$ that I ...
1
vote
0
answers
43
views
Which field of mathematics should be used to formalize the analysis of a dataset's "geometric" structure [closed]
I want to know if mathematical formalism can be applied to understand the "geometric" structure of an experiment's data set, and esp. to the relationships between its categorical predictors.
...
1
vote
1
answer
54
views
Predicted probabilities from logistic vs log-binomial model
I am giving a talk on logistic regression and I was going to mention log-binomial models to estimate risk-ratios. I understand the difference between odds and probabilities and that they only converge ...
0
votes
0
answers
11
views
Principal Component Regression with a Generalized Linear MIXED EFFECTS Model
According to the wiki page on Principal Component Regression, it is possible to transform the beta values obtained from doing regression with PCA data into beta values for the original features. This ...
1
vote
1
answer
55
views
Slope-intercept correlation for Gaussian GLM with log link function
I am fitting an exponential model using GLM regression (assuming Gaussian error and a log link function) to 1000 trials, giving me 1000 slope-intercept pairs that are moderately correlated. I want to ...
2
votes
0
answers
57
views
How do I compare the effect a binary or categorical variable has on a binary target variable between groups?
I have a dataset with three levels:
on the lowest level there are news articles
on the mid level, news articles belong to news outlets
on the highest level, news outlets are grouped by a categorical ...
0
votes
0
answers
31
views
Why aren't residuals in Poisson regression on the scale of the response variable?
I am running models in R using an OLS as well as using a GLM with a Poisson distribution and log link function.
...