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)

2
votes
1answer
20 views

Which glm family to use for ordinal DV?

I'm trying to test whether duration of time spent on the Internet (ratio scale) can predict behavioural problems (ordinal, with scores ranging from 0-10). I just wanted to double check that an ordinal ...
0
votes
0answers
4 views

Can I analyze planned missing (MNAR) data in unbalanced design using SAS PROC MIXED (Generalized LM)?

In my experiment, there are 3 levels of Treatment (A (control), B, and C). All participants (N=109) underwent two of these treatments, the order of which was determined by random assignment to 1 of 4 "...
0
votes
2answers
33 views

Logistic regression with only categorical predictors

So I started off with a model which included both continuous and categorical predictor variables. But now I wanted to drop the only continuous variable (distance to shore), because to my opinion it ...
1
vote
1answer
23 views

How can standard logistic regression model fractional response variable while denominator is available?

I have X and Y variables, as well as a cluster variable (State). X and State are derived from Database A, while Y and State are derived from Database B. X is a sentiment score ranging between -1 and ...
0
votes
0answers
7 views

How to model GEE (generalized estimating equation) in data coming from two datasets?

I would like to model X (sentiment score, continuous between -1 and 1) and Y (smoking status, either 0 or 1). Individuals can be clustered by the "State" variable. It would be the most ideal if I ...
0
votes
0answers
13 views

Glmnet standardize not consistent with prior standardization

I am running glmnet with standardize=TRUE. Now I try to pre-standardize my data, and un-standardize after the regression. But the results are inconsistent: ...
3
votes
1answer
33 views

Link functions for Binomial Regression

So I have a dataset of presence (1) and absence (0) data, but it mainly consists of 0's (~80% of the 5200 observations). Now while constructing my binomial logistic model I am reading (Zuurt et al. ...
0
votes
1answer
17 views

drop1 LRT is zero in R

So for my current binomial model I am dropping some components and I found out that for one variable the results look a bit different. For 'hurseason' (class factor with two levels Y/N), the LRT is ...
0
votes
0answers
17 views

How the distribution affects the calculation of the coefficient in a GLM?

I was wondering how the distribution specified in a GLM changes the coefficients. If I have understood the process, when you fit a GLM, let's say $g(E[Y|X])=X\beta$, most of the software and R ...
0
votes
1answer
28 views

Interpreting dispersion parameters of poisson GLMM with count data

I am working with count data and trying to understand if my model fit is acceptable for this poisson Generalized Linear Mixed Model: Richness.glmer<-glmer(Richness ~ Unit.type + plot.type + (1|NFI....
0
votes
0answers
13 views

What to do when the predicted versus the residuals are clumped in a negative binomial model

I'm working on trying to find out which variables are most important in explaining the amount of research that one country does in another country. The response (the number of studies) is highly ...
0
votes
0answers
7 views

what should be validation parameter for Logistic Regression(LR) in online learning plus rare event scenario?

We have been following below paper to predict CTR( Click probability) of different ad items. This will be used to serve different ads based on probability values. http://olivier.chapelle.cc/pub/...
0
votes
0answers
3 views

RFE (Recursive Feature Elimination) for Poisson Regression with offset [migrated]

It's my first post so I hope I don't make any editing mistakes. Here's my issue : I'm working on count data and am implementing a Poisson Regression with an exposure factor (that needs to go in the "...
1
vote
0answers
12 views

Linear model for testing a ratio of ratios

Our experimental design is as follows: For each of two genotypes (wt and ko), we perform two different gene expression assays (Assay1 and Assay2), and do 4 replicates of each assay. We are interested ...
0
votes
0answers
19 views

Model design and nonconvergence problem with GLMM, incomplete block design in R

I have a two-part question that includes issues with generalized linear mixed models and failure to converge. First, a little bit about my experimental design. I have data where I am trying to test ...
1
vote
0answers
39 views

Book about generalized linear models [duplicate]

Does anyone know a good book about generalized linear models. I am a practitioner and need to master the concepts of generalized linear model, but also my experience tells me that knowing about the ...
0
votes
0answers
43 views

EM parameter estimation for conditional Gaussian model [duplicate]

Let $$X_1\sim N(\mu_{X_1},\sigma_{X_2}^2)$$ $$X_2\sim N(\mu_{X_2}, \sigma_{X_2}^2)$$ where $\mu_{X_2}=c+aX_1$. Also, I have data $D$ (with missing values on $X_1,X_2$). How can I update/estimate the ...
0
votes
0answers
11 views

What is the use of the quantile information on the deviance residuals?

For example, in R we can fit the following glm: model <- glm(formula = am ~ mpg + qsec, data=mtcars, family=binomial) The ...
3
votes
1answer
45 views

Can a generalized linear model use shifted exponential as residual distribution?

I am facing a modeling problem: $t_{ij} = D_i + T_j + \epsilon_{ij}, i=0...641, j\in\mathbb{N}$ where $\epsilon_{ij}$ follows exponential distribution, $\epsilon_{ij} \sim \lambda e^{-\lambda t}, \...
-5
votes
0answers
61 views

Data distribution in GLM [closed]

I want to do glm analysis for my data in r studio, my data are soil data (PH soil, percentage of clay and silt and loam, nitrogen soil), I have 5 plantation and 3 soil depth, plantation and soil depth ...
11
votes
2answers
90 views

Using lm for 2-sample proportion test

I have been using linear models to perform 2-sample proportion tests for a while, but have realized that might not be completely correct. It appears that using a generalized linear model with a ...
0
votes
0answers
19 views

Determining the Repeatability and Reproducibility (Gage R&R) in Minitab within a General Linear Model

I am doing some rather complex data analysis and I was wondering if it was possible to determine the reproducibility and repeatability of a measurement system within a GLM in Minitab. In this data, ...
0
votes
0answers
7 views

significance between comparisions

I hope this is the right forum for this kind of question (otherwise please point me to the right one). We are working on a RNA-Seq data set (10 biological replica for three conditions). We have ...
0
votes
0answers
7 views

Comparing importance of predictors in different datasets in GLM

I want to compare the importance or 'predictive power' of the same feature/covariate in 2 different datasets. Specifically let $[\bf{y}_1,\bf{V}_1]$ be my output & design matrix of dataset 1 & ...
0
votes
1answer
23 views

Goodness of fit: Generalized Linear Models with missing values in R

I am trying to compare two models and check which is the best fit of our data. The R script is below: ...
1
vote
1answer
44 views

Best way to analyse percentage data

I have percentage data and would like to see if these different variables have an affect on certain factors; i.e., I have different habitats of an area e.g., improved grassland: 40%, arable: 15%, ...
0
votes
0answers
8 views

Suggested methods when logistic regression outperforms with boosted outliers

I am using logistic regression to predict binary outcomes with 5 features. When putting 20x weight on the 0.001% outliers the peformance gets a lot better. It seems that some really high/low values ...
0
votes
1answer
62 views

How do distribution functions (e.g. Gaussian, Bernoulli, Poisson, etc.) relate to deep learning?

I know that neural nets use activation functions, but where do distribution functions play into deep neural networks? For example, the h2o.deeplearning() function ...
0
votes
1answer
40 views

MCMC for Probit/Logit model with some 1's flipped to 0's

I would like help constructing a sampler for the following model, which is the latent variable interpretation of either logistic or probit glm (doesn't matter which one to me), with a small twist: ...
0
votes
1answer
27 views

Can I report simple main effects without an interaction effect?

I have a quick question about reporting simple main effects. I am running a two-way mixed ANOVA and the interaction effect was not significant. I understand that if you have no significant interaction ...
0
votes
0answers
27 views

Linear or non-linear model for social interaction with R

The question here is whether the cooperation of people with equal or similar abilities leads to better results than the cooperation of people with different abilities. The setting is a group of ...
0
votes
0answers
20 views

collinearity in conditional logistic regression: glm vs coxph

I am fitting some conditional logistic regression models to wildlife radio telemetry data using a 1:1 paired design, specifically where habitat features at a single telemetry point are compared to ...
1
vote
0answers
6 views

Residuals for zero bounded data

So, I'm running into an interesting problem with my residual plot. For reference, I'm trying to model a response variable "exploration" for a social network. Exploration is the number of people you ...
1
vote
0answers
56 views

Using PCA to model highly correlated variables

Specifically, Andrew Ng states that PCA should be used to speed up algorithms or to visualize data. He also states that using PCA as a way to prevent overfitting is an incorrect application of PCA. ...
0
votes
1answer
38 views

How to improve a regression model without making new measurements? [closed]

Assuming I can't make more measurements, how can I improve a linear model regression? I have the following data: ...
0
votes
0answers
20 views

How to analyse this hierarchical-like data better?

I have a dataset whose structure is as below. Here f1, f2, f3 etc are a set of features. There are differennt models to predict each Vi from all fjs. There are different models to predict each Wi ...
0
votes
0answers
31 views

Negative main effect but positive interaction - GLM in R [duplicate]

I am comparing the density of bird species within towns/cities and the countryside and am interested to see how different habitats affect this. So, I want to see if different habitats (improved ...
0
votes
0answers
3 views

Given these residual and pair plots how should I model the data

I have used the pairs() function in R and here are the results: I used the lm function and the residual plots were obviously ...
0
votes
0answers
18 views

How to add a random effects term to a generalized poisson regression model?

I want to add random effects terms to a generalized poisson regression model. I know this will involve approximation of the loglikelihood (laplace or Gaussian quadratures). Please, i need some more ...
0
votes
0answers
13 views

Can you use the Poisson GLM if there is an upper limit?

I'm analyzing social network data where roughly 10 groups of 100 people are split into different sized teams. (For example, there are 10 schools, but some of the schools have 5 "classrooms" while ...
2
votes
1answer
44 views

Negative binomial regression in R allowing for correlation between dispersion & regression coefficients

In negative binomial regression, the MLE of the dispersion parameter is asymptotically uncorrelated with the MLEs of the regression coefficients (http://pointer.esalq.usp.br/departamentos/lce/arquivos/...
1
vote
1answer
33 views

Specifying Error() in R `aov` function

Consider a data where samples from different populations of 5 species are analyzed after 4 treatments at 3 time intervals. So the independent variables are ...
0
votes
0answers
9 views

GEE returned Mean estimate 0.5, confidence limit 0.5-0.5 and SE of ZERO for a categorical variable

Following were the notes NOTE: PROC GENMOD is modeling the probability that CompositeO='Yes'. NOTE: Algorithm converged. NOTE: Algorithm converged. NOTE: The empirical covariance matrix ...
1
vote
2answers
47 views

Reduce feature levels

I would like to know if someone knows of a way to group the number of levels of a feature that has 100's (even 1000's) of levels to a smaller number of levels - also, what number levels it should ...
-1
votes
0answers
37 views

How to understand the results of generalized linear model in R? [closed]

I am always confused about the results of generalized linear model when I use R package like MASS. For example, we can get the summary below using glm.nb: ...
1
vote
2answers
191 views

Questionable Beta Regression Results

The goal of this regression is to determine whether the amount of leaf disk that an insect consumed varied by what tree the leaf material came from. I'll acknowledge upfront that my coding is rarely ...
0
votes
0answers
21 views

How to calculated test F with only output glm?

I have to calculate the value of the test F (no software needed ) . I read several notes, but I can not understand how I do . The idea is to take Null deviance and Residual Deviance and split in some ...
3
votes
1answer
73 views

What r parameter is used in a negative binomial regression?

From my understanding of the negative binomial regression, we have $Y_i|X_i; \theta$ is distributed $Neg Binom (r_i, p_i)$, where $r_i$ is known and fixed (analogous to a fixed $\sigma^2$ when we ...
1
vote
0answers
22 views

GLM - compare the estimated model with the model with intercept-only

I have this glm output. How do the test to compare the estimated model with the model with intercept-only ? How can I comment on the result ?
2
votes
0answers
15 views

Identify differentially expressed genes across genotypes

I have 40 different genotypes of rice. For each genotype, I have two replicates under control treatment and two replicates under drought treatment. We sequenced those samples and get read counts of ...