Questions tagged [separation]
Separation occurs when some classes of a categorical outcome can be perfectly distinguished by a linear combination of other variables.
168
questions
1
vote
0answers
14 views
Why PCA is not considered in the taxonomy of blind source sepeartion approaches
Blind source separation approaches are divided, in the literature, into four methods, including independent component analysis (ICA), sparse component analysis (SCA), and non-negative matrix ...
0
votes
0answers
13 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 ...
0
votes
0answers
10 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
0answers
17 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 ...
1
vote
1answer
47 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 ...
4
votes
0answers
111 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 ...
0
votes
0answers
15 views
ICA using one receiver
I have a bunch of overlapping time domain signals $s_{i}(t)$ where $i \in \mathbb{Z} \geq 10$ each with a unique period and narrowband frequency support. I want to separate them somehow. I figure I ...
0
votes
0answers
19 views
Interpreting markedly large Logodds in logistic regression? [duplicate]
I am running a logistic regression in R and for some reason, some of my outcome variables o a ridiculously high log odds (ie -14), which corresponds to a p value of 0.9. But this does not make sense ...
0
votes
0answers
72 views
What does nan mean in Logistic Regression output
I made my first model using Logistic Regression in Python. I'm using some dummy variables as predictors, but it resulted all nan value in the summary, except coefficient. Is this meant to be? If not, ...
0
votes
0answers
22 views
On the detection of the predictors responsible for the separation in logistic regression
I'm using multiple glmm models and I have separation and pseudo-separation situations. I want to know if there are any tools in R to extract the predictors ...
0
votes
0answers
17 views
On the test of significance under penalized likelihood estimation
I start using brglm2 package to implement logistic regression under a perfect separation problem. Is there any way to test the significance of the parameters using ...
0
votes
0answers
12 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, ...
0
votes
1answer
18 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 ...
0
votes
0answers
26 views
significance of pvalue for logistic regression [duplicate]
I'm trying to understand how to interpret the significance of logistic regression
If I use the standard iris dataset, the Petal Length is a significant predictor for the Setosa classification. If I ...
3
votes
1answer
162 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
0answers
51 views
How many percentage to randomize and how many iterations in a "what-if analysis"?
I've got complete separated data as such:
...
0
votes
0answers
37 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 ...
19
votes
3answers
947 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 ...
4
votes
2answers
377 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 ...
2
votes
1answer
126 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 ...
1
vote
0answers
75 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) ...
0
votes
0answers
70 views
How to use the kernel trick on a XOR-like dataset
Let's say that I have the following data:
I want to find a transformation of this dataset that will make it linearly separable.
My thought was to bring the data around the origin and then multiply $...
0
votes
0answers
16 views
Perfect separation of a dummy variable with the dependent variable in a binary logistic regression produces high insignificant estimates [duplicate]
I have the following setup:
I have a binary logistic regression analysis with a couple of independent variables. They include a nominal variable that is coded in a dummy form with $k-1$ predictors. It ...
3
votes
1answer
116 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
...
1
vote
0answers
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?
0
votes
0answers
171 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 ...
3
votes
1answer
105 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 ...
3
votes
1answer
113 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 (...
0
votes
0answers
52 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 ...
2
votes
1answer
151 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
1answer
20 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 ...
0
votes
0answers
168 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 ...
2
votes
1answer
279 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. ...
0
votes
1answer
57 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 ...
7
votes
0answers
2k 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 ...
2
votes
1answer
60 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 ...
1
vote
1answer
39 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
1answer
39 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
1answer
52 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 ...
5
votes
1answer
413 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. ...
2
votes
0answers
55 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)
(...
0
votes
0answers
137 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
vote
0answers
55 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....
2
votes
1answer
421 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 ...
1
vote
1answer
100 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 ...
3
votes
1answer
2k 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 ...
0
votes
0answers
36 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 ...
1
vote
0answers
55 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. ...
2
votes
1answer
333 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 ...
0
votes
1answer
48 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 ...