Questions tagged [robust]

Robustness in general refers to a statistic's insensitivity to deviations from its underlying assumptions (Huber and Ronchetti, 2009).

Filter by
Sorted by
Tagged with
1 vote
0 answers
24 views

Why robust standard errors are smaller than OLS errors?

I'm trying to fit some data using a non-linear model. I used nls function in R and everything works as it should. Then I tried with ...
-3 votes
0 answers
12 views

I want to do a robust regression group wise [closed]

I want to do a robust regression group wise.Also, I want to plot the regression for each group separately
  • 1
1 vote
0 answers
19 views

how well would a robust mixed model fit these data? R (rlmer)

I want to investigate Y ~ X1 * X2 + (1|ID on this dataset (there's a plot of these data in that post too, it's the same dataframe) Y is a continuos outcome ...
3 votes
1 answer
62 views

How do Measure "Robustness" in Statistics?

I am an MBA Student taking courses in Statistics. Our prof was comparing two different methods of estimating the parameters for a regression model: General Method of Moments (GMM) and Maximum ...
1 vote
0 answers
77 views
+50

Are "Moments" More Robust Then "MLE"?

I am an MBA Student taking courses in Statistics. We are learning about different ways to estimate the parameters (i.e. coefficients) of a Regression Model. Our professor indicated that there are two ...
1 vote
0 answers
13 views

hard negative mining equivalent for prediction/forecasting problems

Input: past history 3D velocity components of an object up to time t; history size is w (window or lag) seconds in the past. Output: predicted future possible velocities for times in the range t < ...
0 votes
0 answers
6 views

Error structure for fixed effects model

Just theoretically, I am wondering about use cases for cluster-robust standard errors and fixed effects models in conjunction. Let's say I have a panel 10 firms over 20 years and I choose to use fixed ...
0 votes
1 answer
32 views

What's the difference between statsmodels' RLM and robustbase's glmrob?

The Python package statsmodels comes with robust models of linear regression (RLM, https://www.statsmodels.org/stable/rlm.html). ...
  • 1
0 votes
0 answers
15 views

Probit Model When Dependent and Independent Variable Binary

I am running a Linear Probability Model with a binary dependent variable and a binary independent variable of interest. Additionally, I include control variables that can be binary but also continuos ...
  • 1
0 votes
0 answers
30 views

robust::glmRob (method 'mallows') vs robustbase::glmrob (method 'Mqle') - very different estimates

I want to apply a robust logistic regression in R and I tried the functions robust::glmRob (using the method mallows) and ...
  • 71
1 vote
0 answers
26 views

Stopping rule in the rlm() function for MASS package

When I looked at the source code for the rlm() function: ...
  • 251
0 votes
0 answers
40 views

P-Values from Robust Linear Mixed Models?

I'm running a study which is is investigating how different types of smiles (two levels) are processed in different situational contexts (three levels). I currently have a robust linear mixed model, ...
  • 21
1 vote
1 answer
47 views

Running robust statistics: trimmed means or bootstrapping?

I am currently running some mixed ANOVAs for a 2x2 design (two groups, with pre- and post- test). Some of my variable are fine, but some of them violate assumptions for parametric testing, e.g. non-...
0 votes
0 answers
37 views

Linear Mixed Model - Heavy Tailed QQ Plot

I'm running a study in which participants rate the politeness and superiority of two different types of smiles (two levels: rewarding and affiliative) presented in three different situational contexts ...
  • 21
0 votes
0 answers
32 views

How to understand the calculations behind a robust mixed factorial function in R, and the implications of running it? Example provided

In "Discovering Statistics Using IBM SPSS Statistics" Andy Field recommends using Robust versions of existing statistics in situations where not all assumptions are met. I'm running a 2x3(...
0 votes
0 answers
35 views

rlmer error message

I'm trying to run an rlmer model which uses the same random-effects structure as the most parsimonious model derived from a previous simplification procedure with lmer models. The model structure is ...
  • 21
0 votes
0 answers
26 views

How to model a data with heteroskedasticity in variance and outliers in R?

I have a dataset where variance is different across groups, and there exist some outliers too. I used robust linear regression lmrob() to handle outliers. However, ...
1 vote
0 answers
82 views

Difference-in-difference robust to heterogeneous treatment effect - extension of an article

I am trying to extend the results of Gendron-Carrier et al. (2022) article published in the American Economic Journal : Applied Economics which is about the effect of subway opening on pollution. I ...
0 votes
0 answers
25 views

outliers in regression, selection of a specific region of the samples

I have a set of points/samples like the ones in blue in the image below: there is a bunch of wiggly nonsense here and there, and somewhere in the middle the is a region of almost perfect linear fit (...
0 votes
0 answers
19 views

Modifying Variational Inference to be robust to outliers?

Normally, for variational inference, you have some evidence data $Z$, you have some true distribution $P(X|Z)$, and you have a simpler parameterized distribution $Q(X|\theta)$, and you're trying to ...
  • 401
0 votes
1 answer
155 views

Interpreting the output of sppba(), sppbb(), and sppbi() in R

I'm trying to run a robust mixed ANOVA with a between groups factor (2 levels) and a within groups factor (3 levels). I'd ideally like to be able to set and run planned contrasts, have a full set of ...
0 votes
0 answers
35 views

Robust regression after log-log transformation

I’m considering to do a robust regression after the dependent variable and two of the independent variables have undergone a log (ln) transformation. (Because of economic reasons.) The remainder of ...
  • 1
2 votes
0 answers
58 views

Bootstrap-based robustness of a test checking. Is resampling from sample distribution a good procedure?

I aim to check the robustness of 2 groups t-test when samples come from lightly skewed distributions. To approach the problem I though about performing a Montecarlo based robustness analysis using ...
0 votes
2 answers
184 views

The variance of the weighted median and optimal weights

The median $\tilde{\mu}$ of a sample in many ways is analogous to the sample mean $\mu$. Both are an estimate for the population median or mean respectively, and both approach a Gaussian distribution ...
0 votes
1 answer
32 views

Infinitesimal Robustness, influence function of $T$ at $F$

This text is taken from Introduction to robust estimation and hypothesis testing. Wilcox R. First I will write down the description that leads to definition of relative influence on $T(F)$ and then I ...
  • 219
2 votes
2 answers
58 views

Robust distance weighted mean

Given a data sample $\{x_i\}_1^n$, instead of hard omitting outliers by e.g. trimming, one can form a weighted average where we soft penalize observations out in the tails. \begin{align} \mu = \frac{...
0 votes
0 answers
72 views

What is the difference between robustlmm and clubsandwich in R?

Excuse my ignorance, I am trying to get around a problem with my statistics that involves severe outliers issues, with heteroskeskedacity. My model using linear mixed models, in R, with repeated ...
1 vote
1 answer
270 views

Robustness checks for DiD

I am currently looking at a DiD model to evaluate the impact of a policy on ethnic and non ethnic people in the EU. I have done the regression and got the results. I have only two time periods but 10 ...
0 votes
0 answers
42 views

Robust LMM or Random Forest?

I have non-normally distributed, heteroscedacic and autocorrelated data. To overcome these issues I would like to use robust linear mixed models with the robustlmm package in R or random forest ...
  • 83
1 vote
0 answers
110 views

AIC for robust generalized linear models (glm)

How can I calculate Akaike's 'An Information Criterion' with small sample size correction (AICc) for glmrob from robustbase in R?...
  • 76
0 votes
0 answers
45 views

Will models selected as a linear mixed model be the best model for robust modeling?

My question is mostly theoretical but all of my analysis is in R and my models are structured to fit the lme4 package and the ...
1 vote
1 answer
38 views

Multiple Sclerosis: study design to beta test a software

I am a beta-tester of a software that is intended to help the radiologist to interpret MRI reading of Multiple Sclerosis (MS). MS is a disease that, over time, could lead to new lesions, expanding ...
  • 143
0 votes
0 answers
18 views

Significant interaction in robust ANOVA but no significance in traditional ANOVA

In a 2x2 mixed design (2 groups measured at 2 times, n1 = 10, n2 = 12), a subsample violated the assumption of normality (Shapiro Wilk test). Thus, I used robust ANOVA (R-package ...
0 votes
0 answers
41 views

MinCovDet confusion

Consider the following experiment: look at Bitcoin prices for the last several years, then try to estimate the volatility, in the following two ways. Empirical trailing variance (of log returns) over ...
0 votes
0 answers
68 views

minimum determinant covariance matrix and covariance

I am trying to understand minimum determinant covariance. I gather from this stack exchange post that it tries to select a subset of data that is tightly distributed to exclude anomalies, and it does ...
  • 143
1 vote
0 answers
64 views

Reporting robust means and SD to make more comparable to robustbase::glmrob results

I need to report the mean, and SD of two groups for a specific variable, as well as the odds ratio of the two groups using robust logistic regression. The variable does have outliers. I instead want ...
  • 76
0 votes
1 answer
77 views

Why is 50% the best breakdown point for an estimator?

As stated in Wikipedia: Intuitively, we can understand that a breakdown point cannot exceed 50% because if more than half of the observations are contaminated, it is not possible to distinguish ...
  • 1
9 votes
10 answers
3k views

Why is the Median Less Sensitive to Extreme Values Compared to the Mean?

I am sure we have all heard the following argument stated in some way or the other: For a given set of measurements (e.g. heights of students), the mean of these measurements is more "prone"...
  • 6,068
0 votes
3 answers
271 views

Why do I get a lower R-squared value when I add more variables to my multiple linear regression?

I'm using the Regression Learner tool in MATLAB to do robust linear regression on a set of variables. However, with only one variable I get a higher R-squared value than when I'm adding one or two ...
's user avatar
2 votes
1 answer
69 views

Can we report an F statistic on a Robust Linear Model with only Fixed Effects?

I'm estimating a fixed effects model using Linear Regression with the Huber weighting function. I included a boxplot of the data. How do we report an F-value and significance for the entire model like ...
  • 173
0 votes
0 answers
49 views

better to use robust mixed model or mixed model with multiple imputation?

I've tried to run a robust mixed model using multiple imputation, but I can't pool the estimates, as I posted here. If there isn't a way to use rlmer together with <...
  • 71
14 votes
3 answers
2k views

What does it mean for a statistical test to be "robust"?

Is there an intuitive way of understanding what these two sentences mean and why they're true?: "ANOVA is 'robust' to deviations from normality with large samples", and... "ANOVA is '...
  • 777
1 vote
0 answers
39 views

Confidence interval crosses 0 but statistical significance at p < 0.05 using robust post hoc tests, any explanation?

I have 12 groups so 66 contrasts, though I'm only interested in 6 of them, I'm using the mcppb20() function from Rand Wilcox's WRS package in R. I'm using 5000 bootstrap samples as suggested and 0.2 ...
  • 77
3 votes
3 answers
863 views

Why should we use 16-50-84 percentile instead of $\mu$ and $\sigma$ to characterize sampling results such as MCMC

Can someone explain why we should use the 16-50-84 percentile rather than mean and standard deviation to characterize the average and uncertainties of the sampling results, such as MCMC (the Markov ...
2 votes
1 answer
69 views

Question about Huber loss when k=0 in Casella and Berger

In Casella and Berger (page484), the following Huber loss is defined. Then on the next page, Table 10.2.1 shows the Huber estimator for different $k$: In particular, $k=0$ gives the median, which ...
  • 1,898
0 votes
0 answers
19 views

Covariance from time series with asynchronously observed components

Let $(X_t) = (X^1_t, \cdots, X^d_t)$ be a time series in $\mathbb{R}^d$ with covariance matrix $C_t := \big(\mathrm{Cov}[X^i_t, X^j_t]\big)_{i,j}$. Suppose that $X\equiv(X_t)$ satisfies some ...
  • 145
1 vote
0 answers
23 views

Does blind source separation (ICA) work if channels of mixture are observed asynchronously?

Does Independent Component Analysis (ICA - fastICA, SOBI, etc.) work reliably when applied to a multidimensional mixture (observation) $X = (X^1, \cdots, X^d)$ if the different channels $X^i$ of the ...
  • 145
0 votes
0 answers
36 views

Reporting robust regressions?

I'm running robust regressions in R (rlm from the MASS package) for 3 different models of regressions that are all related: DV ~ IV1 + error DV ~ IV1 + IV2 + error DV ~ IV2 + error I was wondering ...
10 votes
2 answers
1k views

Why is maximum likelihood estimator suspectible to outliers?

I'm new to statistics and currently learning abot MLE. Some of the papers I read: Robust Graph Embedding with Noisy Link Weights mentioned MLEs are suspectible to contamination in data, but didn't ...
  • 103
0 votes
0 answers
69 views

Robust Regression Effect Sizes

What are some ways to calculate effect size statistics from a Robust Linear Regression (Specifically one using Bisquare weighting and MM estimates)? I'm trying to find the closest equivalent of a ...
  • 1

1
2 3 4 5
11