Questions tagged [median]
The median is the value below which half the data or probability distribution lies - when the sample size is odd, the median is the 'middle' value of an ordered sample.
779 questions
1
vote
0
answers
31
views
Error propagation through taking median
Let's say I have a small dataset (10-50 datapoints), each datapoint with some uncertainty associated with it, and I take the median. How can I estimate the uncertainty on that median?
Note: I know a ...
0
votes
1
answer
42
views
Advice on deviation method (Mean, STD vs Median, MAD) to detect postural changes
i’m making a rule-based classification program for posture detector through a front camera, webcam (real-time),
it involves a calibration process, it asks the user to sit upright for about 30 seconds, ...
- 1
12
votes
3
answers
526
views
Sign test: p and CI contradiction (significant but CI includes 0 AND not significant but CI doesn't include 0)
I run a sign test in R and get repeatedly contradicting p-values and confidence interval (CI). See for example these results of BSDA::SIGN.test() with two dummy ...
- 566
10
votes
3
answers
653
views
Why is median treatment effect not estimable?
The recent paper Addanki and Bhandari (2024) claims the median treatment effect (MTE) cannot be estimated for treatment and control groups in an observational study.
The authors illustrate with an ...
- 6,644
0
votes
0
answers
31
views
Estimating Median Treatment Effect on matched data from MatchIt package in R?
I've been using the MatchIt package in R to estimate average treatment effect on treated (ATT) for health insurance programs with observational data.
Can I get ...
- 6,644
1
vote
1
answer
66
views
Which measure of central tendency should be used when the question asks for “on average” versus “typically”?
In an exam I came across the following problem:
A company manager recorded the number of employees’ cars already parked when they arrived over nine days:
...
- 347
11
votes
2
answers
786
views
Preservation of mean>median for sums
Suppose $X$ and $Y$ are non-negative real random variables and that the mean is strictly greater than the median for each of the two variables. Does it follow that the mean of $X+Y$ is strictly ...
- 57.1k
0
votes
1
answer
110
views
Test the median of groups with differing sample sizes [closed]
I have a dataset that contains a set of values corresponding to independent groups. I want to to the median for each independent group. The central issue is the number of samples varies for each group....
2
votes
0
answers
45
views
Very robust regression, Median-like, for Infinities
My y variable can have infinities. Fortunately, I am interested in something like a median statistic. Imagine y1= (-Inf, -2, 0, 1, 2). The mean is -Inf, the median is 0.
Now imagine y2= (-1,0,1,1, +...
- 161
3
votes
2
answers
280
views
Interpretation of null hypothesis for t-test on rank-transformed data
Given that a two-sample t-test (e.g., Welch's or Student's) tests for equality of means, and given that the mean of a rank-transformed sample is just the median of the untransformed sample, is it ...
- 41
0
votes
0
answers
46
views
Using Brown-Forsythe tests to check equality in coefficient of variation values in the context of paired experiments design
I want to decide whether the coefficient of variation (i.e., defined as the standard error of estimate/model predicted value in this case) of a modelled parameter is different between two conditions ...
- 101
1
vote
0
answers
83
views
Aggregate Ordinal Data?
In my research, I am examining the impact of AI labels (with vs. without) on various brand perceptions and behavioral intentions. Specifically, I analyze how the stimulus (IV, 4 stimuli in 2 subgroups)...
9
votes
0
answers
290
views
CDC says median number of sexual partners a woman has is 4.3. How can this be?
https://www.cdc.gov/nchs/nsfg/key_statistics/n-keystat.htm
According to the CDC, the median of the number of sexual partners a woman has is 4.3. But how can this be? The median can be 4, or 5, or ...
- 561
1
vote
0
answers
105
views
Insights into median of means estimators in DiD designs
I’m currently working on a research project focused on robust estimators in Difference-in-Differences (DiD) designs, specifically in the classic two-period, two-group setup. My main interest is in ...
- 11
1
vote
0
answers
55
views
Conditional median of $X+Y$ given $Z=z$
I have two independent random variables $X$ and $Y$, where $Y$ is also independent of $Z$, and $E(X\mid Z=z)=0$ for any $z$. We also know that the conditional distribution of $X+Y$ given $Z=z$ is ...
- 305