Questions tagged [separation]
Separation occurs when some classes of a categorical outcome can be perfectly distinguished by a linear combination of other variables.
170
questions
5
votes
3
answers
367
views
Proportion data with number of trials known (and separation?): GLM or beta regression?
I perform a lot of bioassays in which I score mortality not on individuals, but on groups of individuals as a proportion (the denominator, i.e., number of trials, is known):
Sample Data:
...
- 105
3
votes
1
answer
40
views
Unexpected weights in Gradient descent algorithm (linear classification) in python
I am attempting to implement a back propagation algorithm that can efficevley read from a file of features and there targets and predict there outputs correctly, however for sake of testing I am hard ...
- 31
0
votes
0
answers
85
views
Hauck-Donner effect in ordinal generalized regression
I am running a series of ordinal regressions on a rather large data set (n=3640). With some predictors I had issues with violations of parallel lines assumptions, so I decided to run those with a ...
- 1
1
vote
0
answers
19
views
Pooling Profile Penalised LTRs in multiple imputation
I am analysizing data from a clinical trial.
I used multiple imputation to impute the (binary) outcome variable, which is the only variable with missing data.
All of the covariates are categorical and ...
- 11
1
vote
1
answer
80
views
Binomial (1/0) response, Factor explanatory variable: why can't GLM estimate effect when one factor level has all 1s in response?
I have a dataset with a binomial response variable (1/0) and a single explanatory categorical variable with 2 levels.
For one level, the response is all 1s and no 0s (e.g. ...
- 202
0
votes
0
answers
41
views
Is there quasi-seperation here (glmm, logistic regression), and how to take care of it?
I am running a mixed model (logistic regression) using the glmer function of lme4. I have a binary response variable "Germination", one categorical variable "Habitat" with 4 levels ...
- 1
2
votes
2
answers
168
views
For the Logistic model, why is the objective function unbounded below if two sets are linearly seperated?
I am reading Approximate linear discrimination via logistic modeling in the Section 8.6.1 of B & V's Convex Optimization book. On Page 428,
$$
\operatorname{minimize} \ -l(a, b) \tag{8.27}
$$
...
- 23
2
votes
0
answers
66
views
Why PCA is not considered in the taxonomy of blind source separation approaches?
Blind source separation (BSS) approaches are divided in the literature into four methods, including independent component analysis (ICA), sparse component analysis (SCA), and non-negative matrix ...
- 176
0
votes
0
answers
41
views
feature engineering for linear separability and curse of dimensionality
A common method to make otherwise not linear separable datasets separable, is feature engineering - e.g. squaring original feature as show here:
Taken from here. Ultimately, a dataset with more ...
- 1,326
0
votes
0
answers
21
views
How to test (overall) significance of categorical variable(with multiple levels) in bias-reduced binomial-response GLM(brglm package), & posthoc-test?
I have a dataset that has quasi-separation.
My dependent variable is 'Yes' or 'No', and I have two fixed effects and one random effect. The fixed effect 'Treatment' is a categorical variable and has ...
1
vote
0
answers
18
views
Cold Start and First Price Auctions
I have the following contrived scenario... I've participated on various auction platforms where I bid on widgets. Assume that win/loss outcomes on individual platforms are well-separated such that for ...
- 11
1
vote
1
answer
50
views
Is the sum of 3 bits a linearly separable task?
In other words can a linear classifier learn to correctly assign a class (label 0 to 3) for an input of 3 bits? Intuitively this cannot work, since the half-adder circuit contains an XOR block, which ...
- 111
4
votes
0
answers
152
views
Hidden vs Firth vs Shen-Gao logistic regression: dealing with the Hauck-Donner effect
In 1993 a version of penalized logistic regression was introduced by Firth in order to reduce the bias due to outliers and/or (quasi-)perfect prediction in logistic regression: Bias Reduction of ...
- 604
0
votes
0
answers
22
views
On the use of the sampling year as fixed-effect
I have my database in temporal blocks (so I have my occurrence sites into the year of sampling of a few years with maximum 7 years). In order to fix the complete separation problem that I also have, ...
- 135
0
votes
1
answer
28
views
Is there any impact on the results of random forest if we have perfect separation in the database?
I'm having a perfect separation in my database. One of the solutions is to switch randomly some of the values of the dependent variable to the opposite for the variables that cause the problem this ...
- 135
3
votes
1
answer
240
views
Dealing with quasi-complete separation in General Additive Model?
I am modelling influence of fire on occurrence of certain bird species (count response variable) in Before/after control/impact experiment design. I've got intact and burned sites and count data both ...
1
vote
0
answers
56
views
How many percentage to randomize and how many iterations in a "what-if analysis"?
I've got complete separated data as such:
...
- 1,751
0
votes
0
answers
53
views
Options when model complexity and separation causes non-convergence in logistic regression
I have created an example data set here
My data represent the presence/absence of a particular animal species (data$outcome) and measurements of trees. I would like ...
- 677
19
votes
3
answers
1k
views
Analysis of Danish mask study data by Nassim Nicholas Taleb (binomial GLM with complete separation)
Recently, Nassim Nicholas Taleb made this post about the recent Danish mask study, a randomized controlled trial which concluded that the proportions of newly diagnosed coronavirus infections was not ...
- 2,413
4
votes
2
answers
547
views
Are these two 2D datasets the same and can they be separated with the XOR NN?
I'm a very beginner in the neural network topic. So I ran into a problem.
I see on the online lectures:
Famous example of a simple non-linearly separable data set, the XOR problem (Minsky 1969) in the ...
0
votes
0
answers
211
views
Is this logistf approach the right way to deal with complete separation in my logistic regression?
I am performing lots of association tests between genotypes and binary diseases. The genotypes can be very rare, and my tests often have extreme case control imbalance (e.g. 500 controls for every ...
- 133
2
votes
1
answer
172
views
Using Chi-squared test when perfect separation in logistic model?
My dataset is similar to this; it describes the presence/absence of a parasite in six species of animal from two locations. Each row in the dataset represents a different individual animal.
I would ...
- 677
1
vote
0
answers
115
views
penalised logistic regression with brglm - warnings
I've built a logistic model to predict a binary response. I've got four categorical predictors. One of them (posicion) has 6 levels, 3 of which occur not too frequently and are ALWAYS (by definition) ...
- 43
3
votes
1
answer
228
views
Unexpected behaviour of logit regression with glm in R
I recently was puzzled by the behaviour of R's glm when trying to compute a logistic regression
...
- 33
1
vote
0
answers
25
views
Separation and Correlation
In a logistic regression setting, can there be a relation between the fact that covariates are highly correlated and the problem of perfect(quasi-perfect) separation?
- 229
0
votes
0
answers
247
views
What is the cause of this very high standard error in my logistic regression model?
I am running a logistic regression model where the outcome variable is Neurologic Complications, and there are various factors who's impact I am examining. One of the factors (HTN_new1), a categorical ...
- 1
3
votes
1
answer
139
views
Probability of linear separation in a dataset with categorical response
I am simulating datasets to which I fit polytomous logistic regression models. The maximum likelihood estimator of this model is undefined when all categories are linearly separated (and it is quite ...
- 2,161
3
votes
1
answer
267
views
What does it means getting p-values equal to 1 and complete separation in logistic regression? [duplicate]
I was really confused if I should ask this here or in Stackoverflow, but I'll give a shot here. I ran a logistic regression using statsmodels library in Python. However, two things went wrong here (...
- 610
0
votes
0
answers
85
views
Separating hardly distinguishable classes for easier classification
I apologize for this seemingly basic question, but I was just thrust into this data science role and would like to get some advice from some experts.
I have a dataset with a small number of features ...
- 111
2
votes
1
answer
225
views
Logistic Regression assumption
In Logistic Regression, the assumption is that the data must be linearly separable, but if the data is not linearly separable then we can't apply Logistic Regression?
0
votes
1
answer
27
views
Fourier analysis to retrieve components of individual spectra
I have a basic, simple question, I am a physics student, and searching internet gives me a lot of signal processing theory but couldn't find this basic answer, which I plan to implement in my speech ...
- 151
0
votes
0
answers
268
views
Cumulative explained variance between scaled and unscaled data
I was writing a small piece of code to portray the difference between scaled and unscaled data when doing PCA and (as expected) I found that the separability of classes was better by doing the ...
- 111
2
votes
1
answer
465
views
instability of logistic regression
I am reading Introduction to Statistical Learning and it is said (as in other websites) that Logistic Regression is unstable compared to Linear Discriminant Analysis in well separated cases. ...
- 7,575
1
vote
1
answer
83
views
Doubt on d-separation
In the book: Bayesian Networks With Examples in R, the author shows three examples of d-separation:
He cites:
Then, just a few lines below, the author uses the dsep function, which returns FALSE for ...
- 659
8
votes
0
answers
3k
views
Which theorem in Cover's 1965 paper is actually referred to as Cover's Theorem?
Cover's Theorem is stated on Wikipedia (and similarly elsewhere) as
A complex pattern-classification problem, cast in a high-dimensional space nonlinearly, is more likely to be linearly separable ...
- 560
2
votes
1
answer
80
views
likelihood ratio test for logistic regression with a complete separation
Can likelihood ratio test be used for logistic regression in the case of complete separation? I understand that individual parameters can go to infinity so Wald's test is not recommended, but does ...
- 6,126
1
vote
1
answer
42
views
Clustering Proof of Equation
Greetings! Could anyone enlighten me about the validity of this equation? I'm trying to prove it without success. $K$ is the number of clusters, $C_i$ is the $i$-th cluster, $m_i$ is the number of ...
1
vote
1
answer
54
views
Relationship between Cohesion and Separation
Hello! May someone explain to me please how to prove the equation depiicted with the red line? Thank you in advance!
1
vote
1
answer
69
views
Doesn't separation in data affects Random Forest?
I'd like to know if separation in data points given by a certain variable will affect Random Forest in terms of "being a correct built model"? For example, logistic regression is highly vulnerable in ...
- 247
5
votes
1
answer
426
views
Why one result is so wide in this logistic multiple regession
I am doing multiple logistic regression with data with 24 predictor variables and 193 rows.
All predictor variables have values of 0 or 1 and outcome variables (OUTVAR) also has only 2 possibilities. ...
- 8,993
2
votes
0
answers
60
views
Is A ⊥ B | C where one path active but another inactive?
I'm trying to determine if A ⊥ B | C?
I see two paths flowing through elements of C:
(1) B <- C - > A (all variables unobserved, active triple; independence cannot be guaranteed for this path)
(...
- 216
0
votes
0
answers
254
views
Cutting dendrogram at certain point
I have a question about cutting dendrogram like that
It shows some hierarchy in the prison
I have to cut it to separate the group.
Is it possible to cut not all dendrograph at the same height like ...
- 1
1
vote
0
answers
56
views
What is the *formal* definition of separation in *binary* logistic regression?
I am trying to understand complete and quasi complete separation in the context of logistic binary regression. However, I have not found a clear source. I know the seminal paper
https://www.jstor....
- 11
2
votes
1
answer
746
views
Seeking to understand using the Firth correction in Generalized Estimating Equations to deal with quasi-complete separation
In order to deal with complete separation in my data someone suggested that I run penalized GEE (PGEE) by adding a Firth-type penalty term in R.
Although I have read many papers on the Firth ...
- 21
1
vote
1
answer
129
views
Quasi/Complete separation due to huge and infinite values
(R statistics)
My question is regarding this warning. My data contains patients and healthy subjects. Exponential decay is my outcome measure.
I have a example dataset here
I managed to run ...
4
votes
1
answer
4k
views
If logistic regression is a linear classifier why does it fail on linearly separable data?
Logistic regression is a linear model, decision boundary generated is linear.
If the data points are linearly separable, then why does Logistic regression fail? Shouldn't it perform better on data ...
- 143
0
votes
0
answers
37
views
Linear separation in higher dimension [duplicate]
I am having a problem comprehending with the relation of kernel, weight and linear separation.
I have a case where I am given a kernel $k_1$. that has a corresponding mapping $\phi_1$.
And we ...
- 101
1
vote
0
answers
64
views
anomaly detection : check the separability of normal and abnormal data
I'd like to develop an anomaly detection. I have historical data from sensors in the form of time series. The time series can be divided into data of a normal state and data of an abnormal state i.e. ...
- 13
2
votes
1
answer
442
views
Handling quasi-perfect separation in a zero-inflated negative binomial regression in R
I want to run a zero-inflated negative binomial regression in R, but one of my variables exhibits quasi-complete separation and throws errors for both the negative binomial and logistic pieces. I've ...
- 23
0
votes
1
answer
66
views
Dealing with Complete Separation in Logistic Regression when Reporting [duplicate]
I have a question regarding how one would deal with complete separation in logistic regression when reporting the outcome for statistical analysis. For a study, we have group participants into 4 ...