Questions tagged [clustered-standard-errors]
Clustered standard errors represent the version of the general sandwich variance estimator that correct for (potential) grouping of the observations, e.g., repeated measurements clustered within an individual, or individuals clustered within a hierarchy level (geographical region, educational institution, etc.).
218
questions
2
votes
1
answer
24
views
How can I tell if a clutser-randomised crossover trial has made a unit of analysis error?
I am studying the following paper:
https://jamanetwork.com/journals/jama/fullarticle/2698491
This is a cluster randomised control trial with crossover. I want to ensure they have not made a unit of ...
- 333
1
vote
1
answer
31
views
What would be the effect of assuming the wrong homoskedastic/ heteroskedastic spread of data?
If a cluster trial had homoskedastic data, but the regression model used by the authors used 'robust standard errors' intended for use with heteroskedastic data (or vice-versa), would the implications ...
- 333
3
votes
1
answer
36
views
Does the use of "robust standard errors" in cluster randomized trials suggest heterskadistic data, implying there is high between-cluster variability?
Please bear with me. I am only recently familiar with some of these concepts. Please correct any poor assumptions.
I am analysing a cluster randomized trial with crossover between intervention and ...
- 333
0
votes
0
answers
14
views
How to model clustered data structure association when all clusters are observed at the same time?
I have data from many recipes I made that describes how different ingredients subgroups grouped in different parent categories (sweet, sour, bitter, umami ) work together to ultimately lead to scores ...
2
votes
1
answer
25
views
Clustered standard errors using pgmm()
I am estimating a system-GMM model using the pgmm function from the plm package in R:
...
0
votes
0
answers
7
views
Decomposition of variance of difference in means in cluster-randomized trial
I'm learning about cluster-randomized trials and I'm interested in the variance of the difference in means estimator. Can it simply be decomposed between intra and inter cluster variance? How?
- 236
0
votes
1
answer
33
views
How to compute the variance for repeated measurements with clustered observations?
I have some data that looks as follows:
...
- 21
1
vote
0
answers
43
views
Clustered Robust Standard Errors in ZINB
I have a count dataset where a ZINB model was favoured. In my dataset, I have dyads (government-rebel groups) in different armed conflicts. To account for within-group correlation, I want to cluster ...
- 31
2
votes
1
answer
133
views
Derivation of clustered-standard errors
I'm trying to learn about cluster-randomized trials, and I see that people use clustered-standard errors when the unit of analysis is different than the unit of treatment assignment. If not, the false ...
- 236
0
votes
0
answers
17
views
Design Effect in Cluster Randomised Trials
I want to understand why and how variance inflation factor is taken in account with the formula [1+(m-1)*ICC], where M= Size of clusters and ICC is the intracluster correlation.
As I have deduced till ...
0
votes
0
answers
22
views
How to analyze within-subjects repeated measures design with more data for some participants?
I ran a within-subjects study where the study was broken into four time periods and participants were allowed to participate once in each time period. Each time they participated they were exposed to ...
- 113
0
votes
0
answers
9
views
Questions about efficiency of estimators with longitudinal data
I'm modelling data with repeated observations; I'm reading up on options and pitfalls, and have a few questions.
Coefficient estimates are still unbiased and consistent in the presence of ...
- 571
1
vote
0
answers
29
views
Why are FGLS standard errors so low when estimating linear regression on panel data?
I am currently writing my master thesis an the effect of academic publication on anomaly returns. The idea is that in finance literature anomaly - factors/portfolios are proposed. After the factor is ...
- 11
0
votes
0
answers
26
views
Proof of consequences for ignoring dependence in clustered data
This may be a trivial question and apologies if it's been answered elsewhere.
I'm aware that violating the GM assumption of observations being independent, such as with clustered data, can result in ...
3
votes
2
answers
76
views
Time clusters in a randomised trial
In cluster randomised trials, some clusters are treated whereas others are not. Usually, the clusters are geographical unit, e.g. villages. However, am I right that, in principle, the clusters could ...
- 287
0
votes
0
answers
634
views
Clustered standard errors, python statsmodels
I am trying to understand clustered standard errors better.
I created a very simple example: there are 100 rows, first 30 are from group 0 and next 70 are from group 1. I want to compare estimating ...
- 1
0
votes
0
answers
76
views
Clustered data: GEE VS GLM with robust S.E
I am trying to understand the difference between running a GEE with robust S.E., vs running a GLM where robust S.E could be calculated posthoc based on the model output. For example, if one considers ...
- 333
0
votes
0
answers
26
views
Clusters in stepped wedge cluster randomized trial
I want to run a stepped wedge cluster randomised trial (link), over $N \sim 1000$ cities during $M = 4$ weeks. When I analyze the data of the experiment, at which level should I define clusters (for ...
- 236
1
vote
1
answer
25
views
Why is standard error of clustered observations not under-estimated?
I regularly read that clustering will cause standard errors to under-estimated. So I simulated two distributions - one with clustered observations, one without. In both cases the standard error is 0....
- 13.9k
1
vote
0
answers
58
views
Instrument fails when I cluster standard errors
I am replicating the Bhalotra-Figueras paper on women's political agency and health (Paper Here) an IV model where $x$ is measured at the district-level and have an instrument $z$ measured at the ...
- 13
22
votes
3
answers
4k
views
How to fix the problem in this XKCD comic?
Slope Hypothesis Testing (Randall Munroe, xkcd)
The problem with this comic is obviously that the "measurements" are not independent, violating a key assumption for computing valid p-values....
- 1,119
0
votes
0
answers
18
views
Randomised controlled trial stratified by site - inclusion of site as fixed effects or clustered standard errors?
I have a 2-arm RCT which is stratified by two factors:
Site (2 sites)
Disease presence (presence; absence)
The analysis of the primary outcome (binary) will use a log-binomial GLM to generate the RR ...
- 115
3
votes
0
answers
62
views
Clustered vs. GMM-based standard errors: which ones to use in asset pricing?
This question was posted on Quantitative Finance Stack Exchange a while ago. While it was received positively there and generated a reasonable amount of views, no answers have been posted. Thus I am ...
- 64.7k
1
vote
0
answers
84
views
How to conduct a power analysis for a linear regression with clustered standard errors?
I would like to conduct a posthoc power analysis for a linear regression model to see the power of my analysis given the effect size I am finding. However, due to the clustering of standard errors, I ...
- 11
0
votes
0
answers
94
views
F-statistic reported as NA with clustered standard errors
I have a dataset composing of four variables: outcome variable y, explanatory variable x, and id - an individual identification number, and year. I performed a regression with and without clustered ...
0
votes
0
answers
6
views
Testing for cluster dependence with one observation of each cluster (and all its sub-cluster samples)
Given a group of clusters (PSUs) $P$, can we establish their correlation without any repeated observations of the clusters? I have one sample measurement of a set of clusters and am interested in ...
2
votes
0
answers
68
views
Appropriate fixed effect and clustered standard error to use in a diff-in-diff set up
My question is about using fixed effects and clustered standard errors in a difference-in-difference setup.
I am trying to estimate the causal effect of a policy change on a measure of employment over ...
- 21
1
vote
1
answer
44
views
How to estimate cluster robust errors in the presence of fixed effects for groups within cross sectional data?
I am dealing with cross sectional data that has groups within the observations. For example, the dataset of an online firm registered in one country that allows multiple banks from different countries ...
- 205
0
votes
0
answers
115
views
Repeated patients in a matched case control study - how to cluster the standard errors
I'm doing a matched case control study to assess the impact of a disease on healthcare usage. I'm using observational data, and I have a panel of patients with many years observations each. In the ...
- 245
2
votes
0
answers
60
views
Cluster robust standard error
$$\bf{y}=\bf{X\beta}+\bf{\epsilon}$$
In asymptotic setting, assume:
(a) linearity
(b) $\{y_i,\bf{x}_i\}$ ergodic stationarity
(c) weak exogeneity $E(\bf{x}_i\epsilon_i)=0$
(d) Rank condition $E(\bf{x}...
- 145
0
votes
1
answer
262
views
Cluster standard errors on experiment group?
I am running a very simple experiment. There are two different experiment groups, participants are randomly assigned. They have to make a simple binary decision in the experiment.
I have three control ...
- 245
2
votes
0
answers
148
views
Spatial Error Model vs Conley standard errors
What is the difference (pros and cons) of using an OLS with conley standard errors and a spatial error model? Is there a recommendation for when you should use one over the other?
Thanks!
- 21
7
votes
2
answers
1k
views
When are cluster robust standard errors a valid "alternative" to mixed models?
Background
If we have data with multiple measurements per subject, I understand we must model or correct for the non-independence. I believe there are several approaches to doing this. One is a mixed ...
- 499
5
votes
1
answer
257
views
is ab test valid for concluding causal relation, when analysis unit differs from randomization unit?
In a typical A/B test, the randomization unit is user level, sometimes the analysis unit may be page/visit level, like a cluster randomization experiment. In this situation, the iid assumption doesn't ...
- 418
1
vote
0
answers
77
views
How to test if both the mean and variance of the treatment effect is less than the control?
Statistics background
I'm relatively new to experimentation and causal inference. I took up to graduate level stats in school but forget a ton after entering industry.
Situation
I am running an ...
- 11
1
vote
1
answer
339
views
How to estimate disease prevalence using logistic regression when some people are measured several times
I have a binary variable indicating the presence or absence of a disease as well as variables for time and other covariates. SOME (not all) people are observed (tested) several times.
Suppose I want ...
- 499
2
votes
1
answer
242
views
Difference in clustering standard errors of coeftest and sum
Suppose I have a panel data and would like to look at time fixed effects, i.e. effects constant across "state" but varying over time.
My understanding was always, that the estimated ...
- 345
2
votes
0
answers
501
views
R: Panel Data Clustered Standard Errors
I have a panel regression and like to proof the robustness of my model. I estimated a fixed effects model with double clustered SEs.
I estimated:
an OLS - pooled model
Fixed effects model Index: ...
- 21
0
votes
1
answer
91
views
geeglm: Does as.factor() not work with corstr=fixed?
(See "Edits" at the bottom for partial resolution.)
I am interested in generating point estimates for each cluster in my data using geeglm.
However, when I use as.factor() over the ...
- 61
2
votes
1
answer
55
views
What happens statistically, if you create more observations by measuring more aspects of the same observational unit
Let's say that I want to measure the effect of a treatment on the performance of a firm. However my sample is very small. Let's say 10 firms. It is not possible to observe more firms. All these firms ...
- 379
2
votes
1
answer
358
views
Should we always use clustered standard errors in panel data with only two time points, in a multilevel logistic regression?
This is the situation:
I have a binary outcome at two timepoints (T1 and T2);
I'm using a random-effect logistic regression (I mean: just the random intercept, no random slopes) to estimate subject-...
- 1,162
0
votes
1
answer
121
views
can I cluster standard errors within a multilevel model?
I recognise this is similar to this but I couldn't understand the answer to the question and hence am asking again.
Is it possible to use both cluster standard errors and multilevel models together ...
- 121
0
votes
0
answers
2k
views
Different standard errors in Stata and R for instrumental variables (ivreg)
Working with instrumental variables (IV), I noticed differences between reported standard errors in Stata (e.g., using ivreg2) and R (e.g., using AER:ivreg2). In R I found ways to replicate Stata's ...
- 367
0
votes
0
answers
29
views
Which standard errors?
I'm working on a project with panel data and decided to use the fixed effect model, because the hausmann test p-value < 0,05.
No I wonder ho do I know, which standard errors should I use?
There are ...
0
votes
1
answer
126
views
Difference in Difference and clustering errors
I have audience data for weekly TV news broadcasts from 50 different regions. So each broadcast (around 250 in total) has 50 different audience figures. My data contains two years (year1 and year2). ...
- 11
4
votes
2
answers
676
views
Is the Likelihood Ratio test using cluster robust standard errors fixable by Bootstrap (or someting else)?
There is a common agreement about the invalidity of using likelihood ratio tests when computing Maximum Likelihood Estimates (MLE) using clustered corrected standard errors. The main argument is that ...
- 219
1
vote
1
answer
2k
views
Country fixed effects and clustered robust standard errors for logistic regression
I'm wondering if in the same analysis I can use countries as fixed effect and as a cluster for the robust standard errors.
Background: I'm running a multivariate logistic regression with the ...
- 37
2
votes
0
answers
38
views
At what level should I cluster my standard errors and why?
I have a yearly panel data in which each observation is a pair of monitoring stations (stations measuring water quality in rivers) one located upstream and the other downstream, each station in the ...
- 33
1
vote
0
answers
73
views
Cox PH model on 3 groups of same subjects but under varying conditions to see how it affects the survival - should I use clustering?
There's a group of subjects. They experience events depending on (defined by) certain conditions. I was asked by a researcher to perform a sensitivity analysis on 3 different ways to define the ...
- 11
0
votes
0
answers
438
views
Difference between coeftest output: 'twoways' and 'individual' + time dummies
Cross-posted on Stack Overflow: https://stackoverflow.com/questions/68925694/difference-between-coeftest-output-twoways-and-individual-time-dummies
I am currently working with panel data and I want to ...
- 11