Skip to main content

Questions tagged [matching]

Matching refers to a process in experimental design in which observations are sampled in a systematic, non-random fashion to be analyzed more efficiently with special statistical methods.

Filter by
Sorted by
Tagged with
0 votes
1 answer
23 views

What are the differences between PS-match and adjusted Cox regression?

This is more like an extension of the following question: Propensity Score Matching with Cox Regression I am wondering what are the differences between these: matching patients with PS and running ...
Math Avengers's user avatar
0 votes
1 answer
25 views

Primer on Entropy Balancing

I've scoured the internet, but I can't for the life of me, find a comprehensive primer on entropy balancing. I am currently in the process of cleaning data in order to create weights for the part of ...
sverdo's user avatar
  • 1
0 votes
1 answer
34 views

covariate balance loveplot and distance

What is the distance parameter that is plotted on the love.plot ? I am not able to find any reference or explanation for this in the cobalt package. Any help understanding this value "distance&...
Science11's user avatar
  • 577
1 vote
1 answer
42 views

MatchThem and Mixed-Effects Logistic Regression in R

MatchThem and Logistic Regression in R I am conducting a non-randomized study to evaluate the differences in complication rates between two treatments. The observed complication rates are ...
Ludovico Ambrosi's user avatar
0 votes
0 answers
16 views

Inconsistent Matching Results with Matchit After Excluding One Patient

I used the R package Matchit for 1:1 nearest neighbor matching without replacement in my study. Here’s the code I used: R code: ...
user439136's user avatar
0 votes
0 answers
12 views

Matching control group and treatmeant group period in staggered difference-in-differences

I am investigating how different types of electoral systems systems Proportional Representation (PR) or Majoritarian System (MS). influence the level of clientelism in a country. I want to investigate ...
Frederik Valeur's user avatar
1 vote
0 answers
51 views

What should it be if the treatment group is larger than control groups in Propensity-Score-Matching (PSM) -DiD

In an effort to do Difference-in-Differences (DiD), to mimic the Randomized Controlled Trials, we normally use the Propensity Score Matching (PSM) to eliminate the difference between control and ...
Phil Nguyen's user avatar
3 votes
2 answers
72 views

Limitations of propensity score matching

While studying propensity score matching, I was struck by the following thought: When we are running a logistic regression model to estimate $p(Z=1∣X)$ through some form of parametrization and we are ...
richardjoseph's user avatar
1 vote
0 answers
49 views

Statistical analyses after propensity score matching with matchit - Survival analysis- adjustment?

For an analysis of a data set I applied propensity score matching (matchit in R, 5:1, without replacement), which went very well- I dod not lose too much of the sample and got a balanced data set out ...
Kathrin's user avatar
  • 71
0 votes
1 answer
47 views

MatchIt Marginal Effects

I am trying to estimate the ATT of a treatment with MatchIt and am unsure if it is a marginal effect or a conditional effect. In the vignette for estimating the outcome, it says: In addition, with ...
jacob2881's user avatar
0 votes
0 answers
62 views

How to determine the treatment time for untreated in difference-in-differences design

I'm estimating the effect of treatment initiation on health outcomes using a Difference-in-Differences (DiD) design. The timeline is defined relative to the treatment date, not calendar time. For the ...
mar_els's user avatar
0 votes
0 answers
36 views

Coarsened exact matching (CEM) on longitudinal data

I want to compare the effect of 2 educational programs across time. In other words, I have tested 2 groups of students at 3 different time points (T1, T2 and T3). I have performed CEM on the data of ...
CH_1012's user avatar
0 votes
0 answers
20 views

Propensity score matching - How to select observation for matching among repeated measures

I am unsure how to resolve the following problem and am hoping someone might have a suggestion. We have repeated measures data for MS patients and we want to look at potential differences in an ...
LucaS's user avatar
  • 857
0 votes
0 answers
23 views

(Propensity Score) Matching for multiple outcome variables

I have multiple outcomes (dependent variables) of interest. Before estimating the effect of receiving treatment on these multiple outcomes, I want to use propensity score matching as my control and ...
Charles's user avatar
3 votes
1 answer
88 views

How to adress potential issues with interpretation/generalizability after discarding observations in exact matching?

Observations are discarded if they have no exact match, which potentially leads to a different distribution of the covariates in the matched sample compared to the original sample. I assume that at ...
mb1994's user avatar
  • 85
1 vote
1 answer
56 views

Implications of using a caliper when matching: Would it ruin ATE claim?

I've started to learn about matching methods using MatchIt package, and read "Choosing the causal estimand for propensity score analysis of observational studies" (Greifer & Stuart, 2021)...
Farzin Shamloo's user avatar
1 vote
1 answer
56 views

Mahalanobis distance calculation in MatchIt function

I've started to use MatchIt package recently and it's great! I'm also learning a lot by reading the documentation associated with the package. My question is about calculation of Mahalanobis distance. ...
Farzin Shamloo's user avatar
1 vote
1 answer
52 views

Applying weights instead of matching

I want to conduct a Difference-in-Differences (DiD) design study. I have two groups (which you can see in the sector_eu_without variable) that I want to keep equal regarding certain variables: sex, ...
Ronald's user avatar
  • 21
2 votes
2 answers
93 views

Estimating Individual-Level Causal Effects Under Selection on Observables

The value of a randomized controlled trial is that the random assignment of treatment ensures that no confounding is biasing the relationship between treatment and outcome and, therefore, a simple ...
Brian Lookabaugh's user avatar
1 vote
1 answer
40 views

Which kind of matching can i use and why doesnt my propensity score matching work?

I am currently trying to find out if a certain disease characteristic affects long term survival. I have observational data and i am trying to match my cohort, as there are huge baseline differences ...
Joseph Kletzer's user avatar
0 votes
2 answers
140 views

What are the possible solutions for performing full matching on a dataset with a size of 61,000? [closed]

I am working with a dataset of approximately 61,000 observations (with 27,686 treatment and 32,405 controls) and need to perform full matching using the MatchIt ...
eshuns's user avatar
  • 3
0 votes
1 answer
141 views

Use SMD or raw difference in proportions when comparing balance of binary covariates used in propensity matching?

I am a junior in the medical field, and have been tasked with performing a propensity matched analysis of some data, using R. I am looking at comparing the balance of my covariates (both before and ...
sunny_aus's user avatar
0 votes
0 answers
18 views

Matching on actual earnings versus matching on the kind of unexpectedness in earnings

For simplicity, lets assume this is a question about linear modeling, although I am actually looking at some non-linear models and am willing to consider other models if they would be more ...
andrewH's user avatar
  • 3,247
1 vote
1 answer
62 views

How to compare cases and controls after adjusting for age differences?

I have a dataset in which I have different measures (a, b, c) and two independent variables: D which is an independent binary variable representing cases (1) and controls (0), and age is another ...
elven's user avatar
  • 11
2 votes
1 answer
47 views

General question about exact matching on variable/s

I haven't used matching a lot but have certainly read many papers (some quite old and obviously before propensity scores were used much) that have talked about 1:1 matching on age, sex, etc variables ...
LucaS's user avatar
  • 857
0 votes
0 answers
14 views

How to deal with exposure status change and outcome contribution in different exposure status?

I want to estimate the risk of cardiovascular events in patients with diabetes (exposed group) compared to patients without diabetes (unexposed group). How do we deal (in the analysis) with patients ...
Nao's user avatar
  • 101
1 vote
1 answer
23 views

In Matching, are capped confounders a legitimate method of improving balance for matching with highly skewed continuous confounders?

In matching (or similar confounder-control methods such as weighting), are "capped metrics" a "legitimate" method of improving balance for highly-skewed continuous confounders? ...
user11513145's user avatar
1 vote
1 answer
66 views

After matching: How do I interpret the value of the type ‘distance’ (=Propensity score) in the balance measures table of the r-package cobalt bal.tab?

I have used the R-package ‘MatchIt’ to perform (1) a nearest neighbour propensity score matching (NNM) based on the Framingham Heart Study and (2) for comparison, an optimal PS matching (OM) for the ...
user19939387's user avatar
1 vote
0 answers
53 views

When does Matching result in an ATE vs. ATT on observational causal studies?

I have read that matching nearly always yields the $ATT$ effect, but that subclassification matching can yield the $ATE$. I am therefore wondering what is a heuristic for determining what kinds of ...
user321627's user avatar
  • 4,260
1 vote
1 answer
272 views

In propensity score matching, is the estimand to be estimated the ATE or the ATT?

In the propensity score matching literature (Central Role of the Propensity Score by Rubin), the treatment effect estimand is referred to as the "Average Treatment Effect" (ATE). However, in ...
user321627's user avatar
  • 4,260
0 votes
2 answers
63 views

Propensity score matching with near-perfect differentiation

I am working on an analytics task for my job to match control units to treated units to monitor the effectiveness of a new marketing initiative. We decided to use a propensity score matching method ...
Emma's user avatar
  • 1
2 votes
1 answer
42 views

How to compare initial means between weighted cases?

I want to check if the mean value of a certain personality trait differs between young adults in organization X and young adults outside of the organization, who are identical in terms of education ...
MHx01's user avatar
  • 33
1 vote
1 answer
88 views

Is including weights in g-computation not the same as a plug-in doubly robust estimator?

In the R package vignette for WeightIt(), in the section "Modeling the Outcome", it explains that (assuming I'm reading correctly) that the purpose of applying g-computation after creating ...
user11513145's user avatar
3 votes
1 answer
81 views

What are downsides to "genetic matching," particularly outside of causal inference settings?

Multivariate matching methods typically involve two steps. First the user computes $D$, a matrix of the multivariate distances between units. Second, the user applies a matching function (e.g., 1:1 ...
socialscientist's user avatar
1 vote
0 answers
108 views

Assessing whether the probability of being assigned treatment is equal (or reasonably close) between two individuals/groups [closed]

I'm currently studying the textbook Design of Observational Studies, second edition, by Rosenbaum. Chapter 3 Two Simple Models for Observational Studies says the following: 3.1 The Population Before ...
The Pointer's user avatar
  • 2,204
3 votes
1 answer
165 views

What are the main pros and cons of the most commonly used weighting methods?

There are many methods to generate balancing weights in observational studies (see, for example, the many methods implemented in the amazing WeighIt package). I have seen some great discussions about ...
Charly Marie's user avatar
0 votes
0 answers
49 views

When to use paired samples tests? Specific case in clinical research

I am working on the analysis of an observational clinical study. It is basically a control/intervention study, where the intervention group received a novel protocol. In order to assess its ...
LevG's user avatar
  • 303
0 votes
1 answer
82 views

Propensity score matching R swap control and treatment

I am analyzing observational data and want to perform propensity score matching. I would like to compare males and females matched for age,smoking status, bmi, etc. When I coded the male gender as 0 ...
user18802819's user avatar
1 vote
0 answers
22 views

How to construct a common control group in two cohorts with different disease outcomes?

Question details: I have cohort data with a total of 100,000 participants. For some reason, I need to select 2000 participants as controls to construct a control group. I have 3,000 cases of disease A ...
Mou Yuanlin's user avatar
1 vote
0 answers
21 views

How to Evaluate Interaction Effects in Propensity Score-Matched Samples:

Suppose I want to study the association between an exposure X and outcome Y, and I have used the propensity score to match each exposed subject with those unexposed but with similar characteristics ...
zjppdozen's user avatar
  • 347
1 vote
0 answers
23 views

Matching numerical dataset to closest set in library

I have numerical, 1D datasets from measurements that I each want to match to the closest dataset in a library containing similar datasets. Coding is done in Matlab. Unfortunately I am not too ...
Jan's user avatar
  • 11
1 vote
1 answer
52 views

Average treatment effect (ATE) estimation via matching method while outcomes of control population are constant

I want to estimate the average effect of a treatment that was given with a selection bias. To do this, I'd like to use a matching method. Basically, this method involves finding, for each treated ...
HnbBarca's user avatar
0 votes
0 answers
62 views

Propensity score matching - within subsample analysis

I am trying to understand the impact of a graduate degree in mathematics on the wages of students compared to other degrees. I have been running a probability score matching model using a logarithm ...
Giulio Cavallari's user avatar
0 votes
1 answer
42 views

Do I have to use the exact same variables in each step, if I have a two-step propensity score match followed by a regression?

I am using the propensity score match to grow my sample based on a smaller dataset of existing units that received the treatment. The match will find more likely units from its large population that ...
LifelongLearner2's user avatar
2 votes
1 answer
111 views

How to obtain the (weighted or unweighted) L1 imbalance measure for "raw" data

I have two questions regarding the JASA paper, Multivariate Matching Methods That Are Monotonic Imbalance Bounding, by Iacus et al. (2011), the authors produce Figure 2 (left panel) where they compare ...
yungmist's user avatar
1 vote
0 answers
48 views

Is it advised to do treatment-covariate interactions to estimate the average treatment effect when using exact matching?

I am using method = “exact” from the MatchIt package. In the vignette of MatchIt, it is ...
klin's user avatar
  • 11
5 votes
1 answer
38 views

Should I use independent- or propensity score matching?

I work on data form an observational study (n=34) and want to compare the outcome of the individuals to an historic cohort (n=600). Subjects in the observational study received additional treatment ...
Hans Dak's user avatar
1 vote
0 answers
36 views

estimating effect with marginaleffect package

I want to estimate ATE. first of all I used MatchIt package for full matching for propensity score and then I used logistic regression with all of variable in propensity score model after that I used ...
Mahboobeh Taherizadeh's user avatar
1 vote
0 answers
64 views

Estimating treatment effect with G computation

I used the following code to obtain ATE. Before using the 'marginal effect' package, I examined the coefficients of the model and noticed that the 'symptom' variable has a high standard deviation. Is ...
Mahboobeh Taherizadeh's user avatar
0 votes
0 answers
29 views

When to calculate Outcome value in Propensity Score Matching with staggered adoption?

Sorry if this is a basic question (I'm no statistician): In a propensity score matching study, WHEN do you calculate the outcome values (for each treated and control unit) needed to compute the ...
Romain's user avatar
  • 1

1
2 3 4 5
12