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 ...

learn more… | top users | synonyms (1)

0
votes
0answers
8 views

Comparison between different models: hierarchical models and model with dummy

I have these three models: M0<-glm(y=="yes"~x+z+b+c+r+as.factor(w),family="binomial") 'logLik' -1147.734 (df=65) ...
0
votes
0answers
27 views

Using F-test for (generalised) linear models

I am working with regression on a data set and I am looking for a way to compare the results. From the data ($x$) and observed values ($y$) where $y\in[0, 1]$, I have three models: 1 (baseline): ...
1
vote
0answers
21 views

Post-Hoc test on a Poisson regression

Question How can I perform a post-hoc test after a Poisson regression to know which group differ from which other group? Dummy Data Here are some dummy data (coded in ...
0
votes
0answers
6 views

A linear model for testing difference in pairs value between two groups

I have the following experimental design: Values of expression of 3 genes taken from 3 different patients and 3 different controls. R code for generating these data: ...
0
votes
1answer
14 views

nominal variable (fixed actor) only relevant to a subset of data in SPSS / GLM

An example to illustrate the question I have: Suppose Strength is a function of gender (male/female) and, AMONG MALES, testosterone level (low, medium, or high). I don't care about testosterone ...
2
votes
1answer
47 views

Using Poisson GLM for visits to a historical monument - Am I using the right method?

Dependent variable - number of visitors to a historical monument by day Independent variables - Daily average temperature, relative humidity, number of tourists visiting the state by day, etc. My ...
1
vote
1answer
20 views

A dumb question about deviance and saturated models

Deviance is defined as $$ -2(\log L_0 - \log L_s ), $$ where $L_s$ is the log-likelihood of the saturated model. One definition of a saturated model is "a model with a parameter for every observation ...
3
votes
0answers
32 views

Is there a model with additive effects for always positive dependant variable?

When modeling a dependant variable always positive and continuous, models as log-transformed linear model or GLM with log link are generally used. The log-transformed linear model is : ...
0
votes
1answer
25 views

Measure of fit when there is little variation in one variable?

I have two models M1 and M2 that I use to predict Utility. Here's how they look with ...
0
votes
0answers
38 views

When you have a multilevel / mixed effects model, how do you incorporate the random intercepts when making a prediction?

When you have a multilevel / mixed effects model, how do you incorporate the random intercepts when making a prediction? Here is the context: I'm trying to model a Bayesian regression using an index ...
0
votes
1answer
39 views

Best fit with GLM in R

I'm trying to know what is the best GLM fit with this simple dataset and tests in R: ...
1
vote
1answer
32 views

Slight difference in output of SAS proc genmod and R glm

I am trying to reproduce a model fit using SAS proc genmod in R glm and am able to get the same estimates and SE's for all parameters except the intercept and Distance coefficient. SAS: ...
0
votes
0answers
20 views

testing nonlinear hypothesis glm R

I estimate probit model: \begin{align*} P(y=1|x_1, x_2, x_3) = \Phi(\alpha_0 + \alpha_1 x_1 + \alpha_2 x_2+ \alpha_3 x_3) \end{align*} using: ...
1
vote
1answer
39 views

How to choose t-distribution degrees of freedom in “robust” Bayesian linear models

It is well known that in both frequentist and Bayesian linear models, outliers can greatly influence the parameter estimates. Consider the simple example where one outcome variable, $y$, is predicted ...
0
votes
0answers
22 views

A problem with predicted probabilities from a logistic regression [migrated]

I made a logit link, GLM model, in 2 ways. I want to know the chance to win according to these values, for example, for all combinations of these data. If I use expand.grid, I have the same chance ...
3
votes
0answers
16 views

Spatial Autoregressive Poisson model in R

I am estimating a gravity model of migration on cross-sectional data. The Moran I statistic indicates a positive and significant spatial autocorrelation in the residuals of the non-spatial model, and ...
1
vote
0answers
22 views

GLM Model Selection

I have to fit some data to a glm, family=poisson(link="log"). The response variables are X1, X2, X3 and X4. I need an algorithm to fit the best possible model (by lowest AIC). All terms must be ...
0
votes
0answers
17 views

How to conduct a Generalized LM in SPSS, with proportion dependent variable

I must deliver some analysis result report, by using a Generalized Linear Model for a dependent variable I have, which is in proportion values (invisibility), with two factors. The first factor is a ...
0
votes
0answers
15 views

Unable to estimate AIC in R GLM

IM trying to run a simple glm in r. I know my data fits a Poisson distribution so I need to include this in my glm. My model also needs to have ^2 terms in it. When I run this code ...
0
votes
0answers
16 views

How is the intercept calculated in a generalized linear mixed-effects model?

How is the intercept calculated in a generalized linear mixed-effects model? When there are no Random Effects included the intercept is the average of the reference group, but it is not clear to me ...
2
votes
0answers
33 views

GLMs with transformed response variable

I wonder if use of generalized linear models (GLMs) with transformed response variables is correct. My particular case: I compared goodness of fit of several GLMs with response variable transformed ...
1
vote
0answers
16 views

Application of Huber-White Variance Estimates in GLMER

I'm currently working on an analysis in R using GLMER mixed-effects model with a logistic regression framework under the lme4 package. I would like to include empirical (Huber–White sandwich) variance ...
0
votes
0answers
30 views

Do I need a special kind of linear regression for aggregated data?

I have two separate databases on individuals. But these individuals are not both present in the two databases. So I decided to aggregate them into area-level data (such as State-level). One of the ...
1
vote
0answers
19 views

Should I centre or standardise variables in a linear mixed model analysis?

My study is looking at skin lesions in pigs. I have 2 partially cross-classified random factors (weaning pen and finishing pen) and several predictor variables. I have centred pig weights by pen ...
0
votes
0answers
13 views

Generalized Linear Model on SPSS with the 'error': “set to zero because this parameter is redundant”

For my dissertation I have a lot of data and many nominal variables. None of my data is parametric. I tried transforming some of the percentage data with arcsine because it was in proportions and the ...
2
votes
1answer
41 views

How to identify if parameters are estimable, after defining a design matrix

I need to define a design matrix $X$ so that it fits this scenario: For $1\leq i< j \leq 5$, where $i$ are competitors and $j$ are different competitors. Their score is $y_{ij} =$ score of player ...
1
vote
2answers
170 views

What is the distribution of $e=Y-\mathbb{E}(Y)$ where $Y=\exp(u), \ \ \ u\sim\mathbb{N}\left(\mu,\sigma^2\right)$

As $Y$ is log-normal we've $Y\sim \mathbb{LN}\big(\exp(\mu+\sigma^2/2),\exp(2\mu+\sigma)(\exp(\mu^2)-1)\big)$. Now I define $e = Y - \mathbb{E}(Y) = Y - \exp(\mu+\sigma^2/2)$. As $e$ is the ...
0
votes
0answers
8 views

Comparison between the coefficients of two GLMs

I would be happy to have your thoughts on the issue of comparing the beta-values resulting from two different GLMs. Should the beta values be scaled before a statistical comparison? which test would ...
0
votes
0answers
23 views

How much data is needed for GLM?

I'm doing 2 different GLMs. A Poisson GLM for claim counts and a Gamma GLM for claim amounts. I've read a lot of literature about GLMs but I still haven't found a satisfying answer to how much data is ...
0
votes
1answer
47 views

Which model should I use to predict pass/fail scenario?

I am new to predictive modelling. I am unable to choose the correct model for predicting if a student will pass or fail a particular exam. My data set : Input variables: Total_tests_Taken , ...
0
votes
0answers
26 views

How to drop certain values for a factor variable while fitting a GLM? [duplicate]

My response var is a binary variable. In the predictor variable i have a type variable with levels as l1,l2,l3,l4. And when i run a logit (glm(redonse ~ type, family = "binomial"), Some levels of type ...
0
votes
0answers
26 views

substituting zeros in a Gamma regression

I modeled some right skewed data with a Gamma GLM (log link). This is common practice in my field. However, some observations have a value of zero and the Gamma distribution is only defined on the ...
2
votes
1answer
70 views

Fixed/Random Effects GLM for fMRI

As I understand it, this is a fixed-effects GLM (as could be used in analyzing the results of an fMRI experiment): $$Y = X\beta + \epsilon$$ I assume that $Y$ is a matrix of all the data (voxels ...
2
votes
1answer
31 views

How to decide which glm family to use?

I have fish density data that I am trying to compare between several different collection techniques, the data has lots of zeroes, and the histogram looks vaugley appropriate for a poisson ...
0
votes
1answer
26 views

Definition of ordered probit

Can someone please provide a definition of the ordered probit model so that I can calculate the value of $\Pr(y \le k | \mathbf{x})$, i.e. probability of observation falling in the $k$-th class or ...
0
votes
0answers
22 views

Compare different families of generalized models

I have 5 definitive glm models (Poisson, Normal, Lognormal, Quasipoison and Negative binomial) and I want to compare it. data(rwm5yr) Variable response = docvis ...
0
votes
0answers
9 views

Separating zero mean from other means

I have counts as responses to a treatment with several levels including a positive and a negative control. The positive control has a mean value approximately 10 times the 2nd highest mean; while the ...
0
votes
0answers
18 views

Multiple similar dependent variables, each with the same set of idependent variables. How to find a generic model?

I have a number of similar dependent variables $y_i$, $i\in[1;10]$. Each $y$ has a number of corresponding independent variables $x_{ij}$, $j\in[1;20]$. Now I would like to find some generic multiple ...
0
votes
0answers
13 views

Assess the variability of mis-classification errors over many imputed data set

I am conducting logistic regression analysis: The data includes 107 observations, dependent variable is a binary one, there is about 5 covariates which are both continuous, binary and ...
0
votes
0answers
24 views

How to perform Hausman test for lrm() function based models in R

I am working with lrm() function (from mrs package). There is a very nice way to perform Hausman test by using ...
1
vote
0answers
15 views

R - Binary outcome from binary history [closed]

I am new user of R and I am trying to solve an issue: I have a series of 1,0 (which represents data taken in a period) and I would like to predict future with same results (1,0). Better explanation: ...
0
votes
1answer
27 views

P-values vs Coefficient Values in GLM

In the following excerpt, the author noted an association with the dependent variable based solely on the p-value and not the magnitude of the coefficient. Here is a link to the study. I thought ...
3
votes
3answers
257 views

GLM with logit link and Gaussian family?

Can you run a GLM using a logit link with a continuous DV (between 0 and 1)? Generally it's suggested to use a binomial family with a logit link, but I'm guessing that is because the model assumes a ...
3
votes
1answer
39 views

Model selection for random effects: can unselected random effects be used as fixed effects?

I am working on a mixed effects model. What I would consider random effects are year, sampling transect, and sampling location. There are multiple collections taken along each transect, and multiple ...
1
vote
0answers
31 views

Reltionship between “activation function”, “link function” and “family”

I am trying to get an overall picture associated with the glm command in R: It is said that the link function is inverse of activation function. And that the ...
3
votes
1answer
23 views

Where do deviance residuals come from?

I'm trying to understand deviance and deviance residuals using a simple Poisson regression model as an example. Let's say we have a response variable $$ y_i \sim \text{Pois}(\lambda_i)$$ and we assume ...
0
votes
0answers
22 views

Improve the goodness-of-fit of a logistical regression

I've been working on a data set with binary outcome. Logistic regression was used to fit the outcome with several covariates, all of which are categorical variables. I tried to assess the ...
0
votes
0answers
12 views

Natural exponential family and GLMs

In the generalized linear modelling framework, it is often stated that the conditional distribution of the response needs to be in the exponential family (there are some extentions and refinements to ...
0
votes
0answers
15 views

Selecting best model in GLM logistic with AICc, p-values for best are insignificant?

I have one potentially causal predictor and a number of covariates that I tested via AICc model selection in logistic GLM. I found that alone, the causal predictor has a low AICc (~19) and a ...
0
votes
1answer
42 views

Posterior of Logistic Regression

how do we prove that logistic regression has a Bernoulli-distributed posterior? Also, are there any other link functions within the generalized linear model framework that provide a Bernoulli ...