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)

1
vote
1answer
28 views

How is the sigma^2 value (or MSE) for the link function computed in logistic regression in R?

For example, if you have a logistic regression on certain dataset: fit <- glm(y ~ x, data = test, family = "binomial") If you do ...
1
vote
0answers
11 views

Examining trends with interactions and with stratification - obtaining discordant results

I'm examining the effect of income (categorized into quintiles) on a response variable during different years (from 1996 to 2014). I adjust for some other covariates and have repeated measurements on ...
3
votes
0answers
30 views

Modelling flight delays with negative values

Modelling flight delays with negative values I am working on a model to predict whether a flight will be delayed. The data consists of some explanatory variables for flights from a specific airport. ...
1
vote
0answers
17 views

Probability that LM with lesser RSS has greater residual for individual i (or opposite sign)?

You have fitted a basic linear Model #1 (i.e., GLM with identity-link) based on observed data with residuals: $$ Model 1: y_i = \beta_0 + \beta_1 x_{1i} + \beta_2 x_{2i} ... + R_i $$ A colleague ...
0
votes
1answer
11 views

I was expecting 0 and 1 as an answer of a predict function in r

I'm doing a binomial family with method="glm" in train function (caret package) and as result I'm getting predicted numbers like "0.62325028 0.51807017 0.67119878 ..." and I was expecting vector ...
-1
votes
0answers
19 views

Can I use gzlm with Poisson family if counts data are treated as factors? [on hold]

My data called (my data) looks like below and I have 5 DVS(dv1,dv2,dv3,dv4,dv5) note all under one column and three IVS(IV1,IV2,IV3) as a data frame. I keep getting the error below. I am trying to fit ...
0
votes
0answers
23 views

Regression of Higher order [migrated]

I want to fit a model $Y = X_1^2 + X_2^2 + X_1 + X_2 + X_1\cdot X_2$ How to build this in R glm(Y ~ poly(X1,2) * poly(X2,2) how to generalise it to higher order ...
0
votes
0answers
5 views

GLM to asess differences between two treatments with only one IV

I have some data that I've collected in the field while running a trial where some sites (random 50%) received a treatment and the others did not. The dependent variable is the total number of baits ...
0
votes
0answers
7 views

Adjusting for outlier in Fractional logit in R when dv is very small proportion

I used the code from this site: http://stackoverflow.com/questions/19893133/fractional-logit-model-r to estimate a fractional logit model. There are 90 observations in my dataMy dependent variable is ...
0
votes
0answers
24 views

Poisson for percentage data if values are low?

I have percentage data for diet per area (example here.....) I have no data on the individuals contributing to this diet, only for the population as a whole for each area. I want to assess ...
0
votes
0answers
15 views

I used a generalized linear model with multiple dependent variables in R [migrated]

I used a generalized linear model with multiple variables in R .my data (young) looks like below and I have 5 DVS(dv1,dv2,dv3,dv4,dv5) and three IVS(IV1,IV2,IV3) as a data frame. I keep getting the ...
1
vote
1answer
22 views

Model selection: can I compare the AIC from models of count data between linear and poisson models?

I am modeling count data (with offset / exposure parameter). My modeling strategy is use of a Poisson model and a negative binomial regression model. I compare model AICs, which are about -760 for my ...
1
vote
1answer
30 views

Regressing a discrete variable

I have a discrete dependent variable (say, number of units bought) and want to run a linear regression with in-store promotion, seasonality, trend etc. as predictor variables. I'm not sure if it is ...
1
vote
1answer
12 views

Validity of stepwise regression in DistLM

I have a set of nutrient fluxes data and I would like to know which environmental drivers explains the fluxes. I used DistLM and the marginal test showed that none ...
0
votes
0answers
11 views

What is the best test to use for comparing categorical data , both independent and dependent are categorical [closed]

Data is formatted in this way Frequency Treatment 0 CNC 0 CC 1 CNC 2 OC 2 ONC 3 CC 3 ONC
4
votes
1answer
68 views

Likelihood ratio test disagrees with cross-validation results

I have computed two logistic models of the same data (for different formulas) in R, and compared them using likelihood ratio test: ...
0
votes
1answer
67 views

hurdle model with negative binomial distribution of counts - error message and model selection

I'm working with over-dispersed count data, which is zero inflated (~2/3 zeros). I've fit a hurdle model using hurdle from ...
1
vote
0answers
33 views

Strictly positive response in regression: what should my “default” model be?

For unbounded continuous responses, Gaussian errors are the analyst's default model for many reasons, one of them being that their ML estimate coincides with the OLS estimate that has many desirable ...
0
votes
0answers
38 views

Regression with negative skewed data [duplicate]

Sorry I was told that my question previously didn't contain enough detail so I'm trying to reword. I have a large dataset (circa 700 obs) consisting of a continuous response variable and a series of ...
0
votes
0answers
15 views

How does effect() in R calculate error bars from glm?

I have been using glm in R to determine which factors are most likely to predict a positive/negative outcome for a particular problem. Having found the best fit model I was advised to use effect() to ...
0
votes
0answers
15 views

generalized linear model vs. permutation approach to non-normal data

Most of the data I work with are non-normal. Right now I'm using generalized linear models with two factor predictors, and their interaction. I've used this approach for a dataset I'm working on. ...
3
votes
1answer
74 views

R glmer.nb output. How to get $\hat{\theta}$?

I would like to obtain estimated $\theta$ from glmer.nb function in lme4 package. In my understanding this function fits the model: $$ Y_{ij}|\boldsymbol{B}_{i}=\boldsymbol{b}_i \overset{ind.}{\sim} ...
0
votes
0answers
41 views

Banding variables for initial GLM fit, then replacing with continuous versions. Any pitfalls?

I am aware of some of the pitfalls of using banded versions of continuous variables when fitting a GLM (or other model), with a number of discussions on the subject on this forum (and a excellent list ...
1
vote
2answers
26 views

Testing the difference between two parameter estimates in binomial GLM

There are some related posts on this issue, but no answers actually demonstrate the mechanics of how to accomplish the task that I could find. I want to compare two parameter estimates in a binomial ...
1
vote
1answer
44 views

Can one do GLM with LOESS transformed variables

I have binary valued classification variables, and predictors that are not really performing great in GLM with probit/logit model. Some of the predictors are also correlated with each other. I am ...
1
vote
2answers
45 views

Tweedie distribution GLM for manyany() in {mvabund} package

My data follow a Tweedie distribution, and I'm working with multivariate abundances. So I'm trying to use the manyany() function in the ...
1
vote
0answers
45 views

Significance in beta regression and glm binomial

While performing betaregression using betareg R package I noticed that the terms in my model are often significant, even with very small sample sizes. I tried the same model using glm with binomial ...
0
votes
0answers
7 views

effect of accounting clustering

I am analyzing a data where measurements are taken at pups and these pups are nested to their corresponding mom (rats). When I accounted for clustering, factors diet, drug, and interaction of diet ...
0
votes
0answers
35 views

How to find the appropriate family for glm models?

As suggested over at stackoverflow I poste the question here instead: I have a data frame with three variables, where "Resp" is my response variable (count data), F1 is a categorical predictor (4 ...
3
votes
1answer
27 views

What is the canonical link function for a Tweedie GLM?

I was just introduced to the Tweedie distribution (see this or this) but I'm having a hard time finding what the link function is for a Tweedie generalized linear model. Thoughts?
3
votes
1answer
20 views

What family is used in glm for Continuous predictor vs Continuous outcome

I am going to use the glm to estimate nutrient concentration as a function of river flow. My nutrient concentration are not normally distributed and variance is not constant. So, I would like try GLM ...
5
votes
2answers
99 views

Assumptions behind multinomial logistic regression

What are the proper assumptions behind multinomial logistic regression? And what are the best tests to satisfy these assumptions in any statistical software? What are other suitable models, if those ...
0
votes
0answers
24 views

Modelling count data: mean-variance relationship

I have fit a poisson, quasi poisson, and negative binomial model to some count data. To ensure that I have valid models, I am checking that the following assumptions are satisfied: No ...
1
vote
1answer
25 views

transform of predictor variables

Assume I have a linear model like this: $$ y_i = \beta_1 x_{i1} + \cdots + \beta_{ip} x_{ip} + \varepsilon_i \hspace{1cm} i = 1,\dots,n $$ I know that if $y_i$ must be greater than zero, I should ...
0
votes
0answers
28 views

What does intercept means in multiple glm?

I am wondering what exactly means the intercept in the following? Is it a mean? I was told that is the mean of first category. is it? ...
0
votes
0answers
9 views

GLM---Response is binary and independent variables are a mixture of percentages and Euclidean distances

I'm working on a problem to address animal selection of habitat types. I'm using a generalized linear model with my response variable being binary (0 and 1). My independent variables are percentages ...
0
votes
0answers
21 views

Bayesian fixed effects model and invariant variables

Within a fixed effects approach, the effects of invariant variables cannot be estimated. Their effects are captured by the fixed effects. However, when I estimate following Bayesian fixed effects ...
1
vote
0answers
23 views

Bias from stratified sampling

Due to a lack of significance and the large size of the dataset (which had binomial responses with 20,000 responses out of a sample of 15,000,000) my peer has used random sampling to reduce the amount ...
0
votes
1answer
43 views

How to know what function to use to transform $X$ to get the best model fit in logistic regression?

I have simple model $Y~X$ where $Y \sim Bin(1, p)$, so it's a case of simple logistic regression. $X$ is continuous. How to know what function to use to transform $X$ to get the best model fit?
1
vote
1answer
88 views

Is it valid to use a generalized model with no replicates

I've got a designed experiment that is a long-term agricultural station. The field is divided into 11 rows and 6 columns (each column has two sub-columns). Each resulting plot represents a unique ...
0
votes
0answers
20 views

How to determine correct statistical test for repeated measures, differing treatments in same subject experiment

Our hypothesis is that treatment 1 changes our outcome measure by some amount, and that treatment 1 combined with treatment 2 changes our outcome measure by a larger amount. For each subject, the ...
2
votes
1answer
54 views

MATLAB2014b `fitglme` causes error on intermediate results

((This post is a duplicate from Stack Exchange as there was no response there)) MATLAB R2014b's library function fitglme is acting up. It seems to be producing ...
0
votes
1answer
16 views

Generalized linear model - independent variables with many zeros

I am carrying out glms on count data, several of my variables consist of largely of zero values, i was previously told to exclude these variables as it would reduce the model fit. I can't find a ...
0
votes
0answers
29 views

choosing between overdispersed poisson or negative binomial regression

I am performing a GLM on count data (insurance claims) and I wish to compare Overdispersed Poisson Regression (ODP) against Negative Binomial regression. I would know whether there is a practical ...
0
votes
0answers
18 views

Online Learning of GLMs

I am thinking about learning GLMs (well, actually a Zero-Inflated Negative Binomial model) in an online manner. As far as I know, there is no direct way to learn GLMs in an online manner. Therefore, ...
0
votes
0answers
13 views

R package to do a regularized “quasi-binomial” regression

I have data that I want to model with the following data generating process: $y_{i}$~$binomial(p_{i}, N_{i})$ $logit(p_{i}) =\alpha + \beta*X_{i}$ This sort of thing is easily handled in R's glm ...
7
votes
3answers
182 views

Modelling count data where offset variable is 0 for some observations

I'm trying to help a student of a colleague. The student observed and counted bird behaviour (number of calls) in an experimental setup. The number of calls attributable to a specific observed bird ...
1
vote
2answers
70 views

Can I do a t-test to compare t-statistics?

I was trying to fit a 2-level "hierarchical model" all in one go, in MATLAB. But then realised it might be better to do the lower level first, then the higher level. Simply, I have 80 subjects, from ...
0
votes
0answers
8 views

Higher Moments from Factor Models

Suppose that we are fitting a linear factor model to our data $$r=\alpha+Bf+\epsilon$$ where we assume the factors, $f$, are orthogonal. Using this structure we can estimate the mean vector as ...
0
votes
2answers
53 views

When make clusters in a predictive glm model?

If I want to build a predictive glm model, should I make cluster analysis on 100% of observations or on training sample (80%)? Thanks