Multicollinearity means predictor variables are correlated with each other, making it harder to determine the role each of the correlated variables is playing. Mathematically, it means the standard errors are increased. Multicollinearity can have counter-intuitive effects.

learn more… | top users | synonyms (1)

0
votes
0answers
21 views

Co-linear covariate in regression what if take one out?

I understand that there are a few ways to deal with correlated or co-linear covariates ( PLS or AIC, and Lasso) But my problem here is: If you have 2 covariates x1 and x2 that are correlated(you ...
0
votes
0answers
39 views

Edited: Glmer model won't run all factor levels due to multicolinearity - can you have pseudo-multicolinearity?

I have a repeated measures design in which I have tested 112 individuals each with three different 20 second playback stimuli. I used a random block design where an individual was assigned to 1 of 6 ...
1
vote
1answer
24 views

transform variables then check for collinearity or other way round?

I was planning to construct a model with around 19 explanatory variables and one response variable. I have some confusion on following thing: If transformation is required in case of explanatory ...
2
votes
1answer
39 views

Multiple regression with correlated variables

I am trying to do multiple regression analysis of a continuous health variable (yval) keeping age, gender, height (cm), weight (kg) and waist (cm) as predictor variables for a database of about 7000 ...
0
votes
1answer
24 views

What to make of countervailing spatial regression coefficients?

I am running regressions across a country's counties (N about 300). I divide the country in two regions A and B to control for potential unobservables. My explanatory variable varies at the county ...
3
votes
1answer
36 views

Principal Component Regression with an additional factor

I am looking to tease out the significance and contribution of a particular variable to 2 different continuous responses. I have 7 continuous variables I know to be influential on the two responses ...
1
vote
0answers
7 views

Three percentage indicators into one (or two measures) in regression moder

I am working with a data set resembling the extract below: ...
2
votes
1answer
44 views

Invertibility of $X^TX$ with severe multicollinearity in regression

I am studying about multicollinearity in regression and in the book it says, "if there is severe (but not perfect) multicollinearity, two or more predictor variables are highly correlated, so $X^TX$ ...
0
votes
0answers
12 views

What happens under the coexistence of biased estimation and multicollinearity?

I understand that multicollinearity induces inflated standard errors but doesnt bias coefficients while error-correlated omitted variables biases coefficient but what happens under the coexistence of ...
3
votes
2answers
48 views

Collinearity in multivariate regression with huge amounts of data

Take the following example. I wish to predict physical performance as a function of height and weight. I already know weight negatively affects performance. Height also negatively affects performance, ...
0
votes
0answers
13 views

high correlation between IVs, but VIF and tolerance normal?

In a multiple regression analysis: If two variables are correlated by more than .8, but the VIF for each of them are 1.005 and 1.006 and the tolerance numbers .994 and .995, what exactly does this ...
0
votes
1answer
93 views

Should we test error terms for auto correlation or multicollinearity

I understand the basic difference in definition between multicollinearity and autocorrelation. I.e multicollinearity describes a linear relationship between whereas autocorrelation describes ...
0
votes
0answers
22 views

Linear Regression - Number of Predictors - Negative Weight [duplicate]

I am trying to prep myself for data science interview & I saw the following question on a forum You have several variables that are positively correlated with your response, and you think ...
3
votes
0answers
45 views

Variance Inflation Factor less than 1 in ridge regression?

I was trying to determine the biasing constant in ridge regression when I came across a phenonomenon that seems quite puzzling, to me at least. I let the GCV criterion choose a constant for me and ...
3
votes
2answers
99 views

Predictor variables sum up to 1 but not necessarily correlated - is it a problem? [closed]

I am trying to fit hierarchical mixture model (using ML and MCMC, but this shouldn't matter) where the linear predictor part contains 17 independent variables. These are habitat variables: for each ...
1
vote
0answers
36 views

Principal component regression on polynomial terms

One of the data sets I am working upon had 3 variables which were having almost 100% correlation among themselves. Since I am learning regression modelling I thought I'll do principal component ...
0
votes
0answers
29 views

Multicollinearity? Weighted or unweighted covariance?

I am working with several regression models with different dependent variables where two of my independent variables of interest are correlated: I am wondering if the large standard errors associated ...
0
votes
0answers
7 views

What is the best way to test if an IV has a stronger effect on the DV than another IV that a subset of the first IV?

I wish to investigate the effect of general international experience of a corporation measured as the number of foreign subsidiaries it own vs. the effect of local experience in a region measured as ...
2
votes
0answers
66 views

Should fixed effect dummies be included in multicollinearity diagnostics (VIF calculations) for regression?

I am currently in the progress of performing multicollinearity diagnostics for a logistic regression model using tolerance and VIF calculations based on recommendations in Allison (2012) (Logistic ...
5
votes
1answer
75 views

Why do we use the term multicollinearity, when the vectors representing two variables are never truly collinear?

When two vectors $a$ and $b$ are collinear, then $a = xb$, (where $x$ is a scalar) so in linear algebra, collinearity is a narrowly and clearly defined (and binary) concept. Two vectors -- in my ...
1
vote
0answers
13 views

Variable selection multiple data types

I'm working on a data set containing 45 predictors that cover metric, ordinal and nominal variables. Aiming to finally build a model, which explains the interacting effect of a set of predictor ...
1
vote
1answer
172 views

what if response variable is 'yes or no' in R?

How to analyze above the data to predict the probability that people have disease with a model? Factors thought to influence infection include city, age, and diet. BUT, I don't know how to do ...
2
votes
0answers
50 views

Logistic principal component regression where PCs are correlated with an additional binary predictor

My scenario is this: I collected a bunch of vegetation data (% cover counts in a quadrant at different heights) in patches where birds were seen foraging and also in control patches where no foraging ...
2
votes
1answer
19 views

Covariate related to independent variable - best solutions

I have 2 groups (tinnitus sufferers and controls) who are significantly different in age - I would normally control for age as a covariate (as it is a cognitive task) but it violates the assumption of ...
0
votes
0answers
57 views

Multicollinearity test for panel data

I am using the within estimator (R, plm package) on an unbalanced panel with stock returns as the dependent variable and a number of financial ratios as the independent ones. I want to test for ...
0
votes
1answer
61 views

About stepwise regression and correlation

I am trying to fit a model to some observed data with the least squares method. Now, I am at the stage where I have run a stepwise regression (traditional), with Entry level $=0.025$ and Stay level ...
1
vote
1answer
28 views

Linear regression with redundant features (perfect multicolinearity)

Suppose $X \sim N(0,1)$, $Z=X$, and $Y=X$. An ordinary least squares regression problem is solved: $min_{(b1,b2)} \|Y-(b1*X+b_2*Z)\|_{2}^2$ This is a strictly convex function which must have a ...
1
vote
2answers
70 views

Multicollinearity in multiple regression

I really hope you can help! I'm in the last stages of my PhD. My supervisor is keen on including all variables in the multiple regressions I am running. Some of the scales are intercorrelated (some ...
1
vote
0answers
6 views

Correlation between offset and predictor in count data

I am fitting a Poisson model that requires an offset term to account for sampling effort. However the offset term is highly correlated with one of the linear predictors that is central to my ...
0
votes
0answers
8 views

Coefficients flip sign in general linear model depending on what predictors are included: collinearity is NOT a problem [duplicate]

I have a general linear model with several predictors (~10). The sign (beta) of one of the predictors (Pred1) is negative when all predictors are included. It's STILL negative when the most correlated ...
0
votes
0answers
48 views

Combining collinear continuous predictor variables in GBM

I'm dealing with a dataset (n=254) with one dependent variable (Y), and three independent variables (X1, X2, X3), all continuous. I would like to compare the contribution from each IV to Y. I've been ...
1
vote
0answers
15 views

Centering my variables decreases the condition indices

In my regression model, one of my condition indices is very high, but only the intercept and one of the eplanatory variables have a large variance decomposition proportion for this index. The VIF is ...
2
votes
0answers
54 views

Very low VIF values, but extreme high condition index

In my multiple linear regression model, all of my explanatory variables have a VIF score, lower then 3, but the highest condition index is 709. The constant and one of the explanatory variables have ...
0
votes
0answers
38 views

Regarding analysis of regression result and vif result

I am working on building a regression model. There are 51 points. The number of predictor variables is 37. The following is the result of running lm result. When trying to detecting the ...
0
votes
0answers
24 views

How Can I Build A Regression Model With Collinear Data?

Hello there my fellow Cross Validated members; I’m here today to brainstorm a little bit with all of you out there, to flesh out our collectively acquired data analytic skills, and to try and find new ...
0
votes
0answers
25 views

nonsignificant predictor becomes significant

This may be a question with an obvious answer. I am trying to predict change in a continuous variable over time using a linear mixed model in SPSS. There are 4 time points. Ultimately, I would like to ...
0
votes
0answers
10 views

Substantial changes in significance level when adding more variables to the model [duplicate]

I have a multiple regression model. When I add one more independent variable to the model the significance level of two of my original independent variables suddenly get insignificant. How come? All ...
1
vote
0answers
17 views

Multiple Reg with 2 Independent Variables that are Correlated - Orthogonalizing the IV's

I have two Ind. V's, $x_1$ and $x_2$. They are slightly correlated with eachother. $x_1$ explains a significant portion of $y$'s variability. Rather than just modeling $y = \beta_0 +\beta_1 x_1 ...
0
votes
0answers
28 views

Multicollinearity, feature selection for discriminant analysis and the error rate

I have a question regarding feature selection in LDA/QDA and deciding to eliminate variables to find an optimal model (lowest misclassification rate) I'm looking at how quadratic and linear ...
2
votes
0answers
75 views

Looking for and dealing with collinearity in a GLM

I've got this dataset with one continuous dependent variable and two categorical explanatory variables. I'm wanting to run glms on the data but I'm finding problems with what I think is collinearity. ...
0
votes
0answers
21 views

Mean centering variables NOT in a moderation/interaction term [duplicate]

I am trying to assess the impact of multicollinearity in a regression because I have two separately measured variables which have the reversed signs problem (one predictor is +b regression weight, the ...
1
vote
1answer
80 views

Multicollinearity among categorical variables - Is it normal?

I'm building a logistic regression model in which almost all of the input variables are categorical. There are multiple sets of categorical variables, for example, day of the week, age range buckets, ...
3
votes
3answers
70 views

Too many predictors to manually check linearity

Say I have 1000 potential predictors in a logistic regression. I don't have time to check each predictor manually for linearity. I could wait till after variable selection, but in that case I wonder ...
0
votes
0answers
22 views

Regression on a Ratio with Numerator and Denominator as regressors

Assume we have a dataset with prices of train-rides. There is the price for the ticket, the distance of the ride und some other relevant variables (e.g. x2: 1st/2nd class, x3: name of train-company ...
2
votes
0answers
83 views

Is multicollinearity a problem with gradient boosted trees (i.e. GBM)?

A question about multicollinearity for random forests has been asked and answered, but what about boosted trees?
0
votes
1answer
46 views

Curse of dimensionality mimics multicollinearity?

Why does the curse of dimensionality mimic multicollinearity, in the following sense.. Consider the random vector $Y = [y_{1}, \dots, y_{4}]$ where each element is ~ Uniform (0,1). Take 10 samples ...
4
votes
2answers
131 views

Many dependent variables, few samples: is this an example of “large $p$, small $n$” problem?

"Large $p$, small $n$" typically refers to "many independent variables, few samples". In my case, I have $1$ independent variable, $300$ dependent variables, and $n < 20$ samples. Thus, my case ...
0
votes
0answers
23 views

Regularization vs PLS for highly colinear data?

When dealing with colinear data, when would I want to use L1/L2 regularization, and when would I want to use Partial Least Squares? Are there any theoretical or practical reasons to prefer one over ...
0
votes
0answers
36 views

vif output for multicolinearity when you have factors

using vif(model) in R provides us with the variance inflation factors for out model. often people use sqrt(vif(model)) > 2 ...
2
votes
1answer
149 views

How to evaluate collinearity or correlation of predictors in logistic regression?

In linear regression it is possible to render predictors insignificant due to multicollinearity, as discussed in this question: How can a regression be significant yet all predictors be ...