Questions tagged [exchangeability]

A set of random variables is exchangeable when their joint distribution is invariant under any permutation of the random variables.

Filter by
Sorted by
Tagged with
6 votes
1 answer
166 views

Is the converse of de Finetti's theorem true?

As I understand it from Wikipedia, de Finetti's theorem says: "Exchangeability implies that variables are conditionally independent given some latent variable". Is the converse true as well?
0 votes
0 answers
12 views

Multivariate Hypergeometric distribution for non-exchangeable sequence

I have a problem that I think is similar to the Multivariate Hypergeometric distribution urn problem as described on Wikipedia. I have a dataset where each row is a trial. All trials have the same ...
  • 101
2 votes
2 answers
91 views

Show that the variance of the longitudinal estimate is $ \dfrac{2 \sigma^2(1-\rho)} { n} $ rather than $\dfrac{\sigma^2(\rho)}{n}$

patients are randomızed to elther treatment, their pressures are measured at baseline, treatment is administered for two weeks and $\mathrm{BP}$ is then measured a second time. The treatment effect is ...
user avatar
3 votes
0 answers
28 views

Are there examples of ML or stats approaches that are valid for IID data, but not exchangeable data?

Lots of supervised learning theory is motivated using the IID assumption. Do most of these methods apply equally well if data is only exchangeable, and not IID? Can you provide an example where this ...
1 vote
0 answers
71 views

How to derive g-computation for a longitudinal experiment from sequential conditional exchangeability?

We have a longitudinal experiment, with interventions $\bar{A}=\{A_1,A_2,\ldots,A_K\}$ and outcomes $\bar{Y}=\{Y_1,Y_2,\ldots,Y_K\}$. Using some sequential conditional exchangeability assumptions, it ...
  • 376
4 votes
2 answers
262 views

Switching interventions after randomization

Suppose in a randomized control study, we assigned each individual to one of two interventions (A or B) with equal probability. And suppose for the sake of argument, there is complete balance in ...
  • 551
11 votes
3 answers
844 views

How do you calculate the expected value of a discrete distribution without replacement?

Say I have a set of 10 values I want to draw 3 values from, uniformly, without replacement. For instance: $$S = \{0,0,0,0,22.95,0,0,0,19.125,25.5\}$$ With replacement, this seems simple, you just add ...
  • 223
2 votes
0 answers
184 views

Exchangeability, causal inference, and partial pooling

In Statistical Rethinking, Richard McElreath writes the following concerning the use of partial pooling (i.e. varying/random effects) in Bayesian hierarchical models: Could we also use partial ...
  • 139
1 vote
0 answers
13 views

Proof for a seemingly simple property for fully-connected coloured random graphs?

I have a probability distribution defined over a set of fully-connected simple graphs depending on their coloring. Let us have a fully connected graph with $N$ nodes, a node may have a color $i$ ...
  • 151
0 votes
0 answers
35 views

Proving Independence due to exchangeability?

I have a set of bernoulli random variables $\{x_i\}^{n}_{i=1}$ and $\{x_{ij}\}_{i< j}$. They have a probability distribution with following conditional independence: $$P(\{x_i\}^{n}_{i=1},\{x_{ij}\}...
  • 151
8 votes
1 answer
910 views

Difference between exchangeability and independence in causal inference

When inferring causal effects from observational studies, one of the assumptions that's generally required is the exchangeability assumption. Suppose $A \in \{0, 1\}$ is a binary treatment, and let $Y^...
  • 331
2 votes
0 answers
146 views

Conditional exchangeability when conditioning on continuous variable

Conditional exchangeability is often introduced in a simple setting with a binary outcome, binary treatment, and binary confounding variable. In this setting, exchangeability holds within strata of ...
1 vote
0 answers
233 views

Practical example for (conditional) ignorability in potential outcomes

Within the potential outcomes framework (Rubin Causal Model), I find that most sources efficiently explain ignorability and conditional ignorability algebraically. As far as I can tell, some use "...
1 vote
1 answer
123 views

correcting t-test for dependent but not paired data (in R)

My data $\{X_1,...X_N \}$ are assumed exchangeable with known correlation $\rho$. I want a "t-test" (or similar) of mean $E(\bar{X})=0$ in the R language. More generally, i seek an easy and ...
  • 133
3 votes
0 answers
34 views

Expectation of uniform variates

Let $X_{1},X_{2},X_{3}$ be random variates from $U(0,1)$. It is required to compute $E(\frac{X_{1}+X_{2}}{X_{1}+X_{2} + X_{3}})$. Here is what I did.. $E(\frac{X_{1}+X_{2}}{X_{1}+X_{2} + X_{3}}) = E(1 ...
  • 1,040
1 vote
0 answers
125 views

Total correlation (mutual information) of exchangeable variables

The total correlation (aka multivariate mutual information) between a bunch of random variables $X_1,...,X_n \in \mathcal{X}_1 \times ... \times \mathcal{X}_n$ with joint density $p(x_1,...,x_n)$ and ...
  • 266
1 vote
1 answer
255 views

Estimating and testing correlation of longitudinal random variables

Each patient (indexed by $i$) contains multiple measurements of two variables $X_{i,t}$ and $Y_{i,t}$ over time $t=1, \dots, T$. For each time point $t'$, assume the correlation $\mathrm{cor}(X_{i,t'},...
3 votes
0 answers
134 views

understanding exchangeability in GP regression

It is well known that exchangeability refers to the following property $p(X_1,\dots, X_n) = p(X_{\pi(1)}, \dots, X_{\pi(n)})$ for any finite $n$ and a permutation $\pi$ when we have an infinite ...
6 votes
2 answers
514 views

Exchangeability and joint distribution

The definition of an exchangeabilty for a finite sequence says that, if we have random variables $X_1,\ldots,X_n$, then for each permutation $\pi: \{1,\ldots,n\}\rightarrow\{1,\ldots,n\}$, the joint ...
8 votes
3 answers
1k views

Permutation tests and exchangeability [duplicate]

Permutation tests assume exchangeability of the response/observations under the null hypothesis. In what practical situations is this clearly violated? When is it unproblematic? Edit/additional ...
  • 11k
0 votes
1 answer
149 views

How does one actually determine if a sequence is exchangeable

Pardon me if the question is stupid, but I don't understand how you go about proving exchangeability of a random sequence. In the case of a distribution such as the normal distribution, I get that ...
  • 1
2 votes
1 answer
204 views

Can We Use Permutation Test Everywhere to Check Difference When the Null Hypothesis Is "There Is No Difference Between Two Groups of Data?"

What I Am Trying to Do? Let's assume, I am working on a research problem where only one participant is present. I have a data set consisting that participant's respiration rate (The number of breaths ...
6 votes
5 answers
393 views

Is independence subjective?

I am trying to better understand exchangeability. Suppose that I would like to run an experiment. I will pick two people and give each a fair coin and tell them to toss N times in a row. The ...
4 votes
3 answers
845 views

Relationship between exchangeability and independence

If the random variables $X_1,...,X_n$ are I.I.d. when conditioned on an unknown parameter $\theta$, where $p(\theta)$ represents our prior beliefs about $\theta$, it follows from the commutativity of ...
4 votes
0 answers
465 views

Intuitive understanding of the Aldous-Hoover representation theorem for row-column exchangeable arrays

I would like to ask a couple of questions about the Aldous-Hoover theorem for the representation of probability distributions over (2D) arrays with exchangeable rows and columns. I'd be happy about ...
  • 1,286
1 vote
1 answer
592 views

What does it mean "a conditional expectation given a stochastic process"?

Let $(\Omega,\mathscr{F},P)$ be a probability space. Let $X,Y$ be random variables on $\Omega$. Then, we say $Z\sim X|Y$ iff (i) $\int_{Y^{-1}(A)} X dP = \int_{Y^{-1}(A)} Z dP$ and (ii) $Z$ is $\...
  • 131
3 votes
2 answers
313 views

Permutation test comparing nested non-linear models with an exchangeable dummy variable

This question is closely related to an earlier question, but I realized my case was actually a lot more specific than the way I formulated it there, in ways that I think merit a separate answer. I ...
2 votes
0 answers
95 views

Are random variables sampled upon stopping rules exchangeable?

In this article from D. Berry https://www.jstor.org/stable/2684222?seq=1#page_scan_tab_contents the author uses an example to introduce some limits of p-values in frequentists analyses. He takes an ...
  • 3,063
4 votes
0 answers
313 views

Correlated Bernoulli Trials

Suppose there are $n$ dependent Bernoulli trials, $X_{1}$,...,$X_{n}$ with $% X_{j}\in \{1,0\}$ and $\Pr (X_{j}=1)=q$ for all $j=1,...,n$. For any $% n\geqslant 2$ dependent Bernoulli trials, in the ...
1 vote
0 answers
44 views

Can I predict if my sequence is exchangeable? [closed]

Supppose I have a sequence as so (which we can really think of as a sequence of smaller sequences): [[1 1 1 1 1 1], [2 2 2 2 2 2], [3 3 3 3 3 3], [4 4 4 4 4 4]] ...
  • 857
3 votes
1 answer
170 views

When $(X_1-X_0, X_1-X_2)\sim (X_2-X_0, X_2-X_1)\sim(X_0-X_1, X_0-X_2)$?

Consider a bivariate distribution function $P: \mathbb{R}^2\rightarrow [0,1]$. I have the following question: Are there necessary and sufficient conditions on $P$ (or on its marginals) ensuring that $$...
  • 538
7 votes
2 answers
273 views

Correlation between an observation and its rank in a random sample

Suppose $X_1,X_2,\ldots,X_n$ are i.i.d random variables with an absolutely continuous distribution. We say the observation $X_i$ has rank $R_i$ if $$X_i=X_{(R_i)}\quad,\,i=1,2,\ldots,n,$$ where $X_{(...
  • 9,641
2 votes
0 answers
59 views

How is exchangeability related to covariate shift?

I understand that exchangeability refers to the notion that the order of data in a sequence does not affect the joint distribution of that data. In a sense, the current data we possess is from the ...
15 votes
4 answers
647 views

If $X_1,\cdots,X_n \sim \mathcal{N}(\mu, 1)$ are IID, then compute $\mathbb{E}\left( X_1 \mid T \right)$, where $T = \sum_i X_i$

Question If $X_1,\cdots,X_n \sim \mathcal{N}(\mu, 1)$ are IID, then compute $\mathbb{E}\left( X_1 \mid T \right)$, where $T = \sum_i X_i$. Attempt: Please check if the below is correct. Let ...
  • 283
3 votes
0 answers
103 views

From bivariate to trivariate probability distribution

Let $\mathcal{G}$ be the space of all possible bivariate probability distributions. Let's pick a bivariate probability distribution $g\in \mathcal{G}$. Can we always find a random vector $(X,Y,Z)$ ...
  • 538
8 votes
2 answers
252 views

Are products of exchangeable RVs exchangeable?

Assume that $$X=(X_1, ..., X_n),: (\Omega, A,P)\to (\{0,1\}^n, 2^{{\{0,1\}}^n})$$ and $$Y=(Y_1, ..., Y_n):(\Omega, A,P)\to (\{0,1\}^n, 2^{{\{0,1\}}^n})$$are two random Variables that have binary RVs ...
  • 2,781
2 votes
1 answer
764 views

Sampling from a finite sequence without replacement yields exchangeable sequences?

I have read (and re-read) the wikipedia article on Exchangeability https://en.wikipedia.org/wiki/Exchangeable_random_variables . The disconnect for me is that : after having sampled without ...
3 votes
1 answer
133 views

Can every parameter $\Theta$ in Bayesian modelling be explained via De Finetti`s representation theorem

My question is the following: I recently got to know (and love) De Finetti`s representation theorem and I now started to read a Book an Bayesian statistics. However this book simply takes as the ...
  • 2,781
5 votes
1 answer
406 views

Wilcoxon signed rank test independence assumption

Let us say we perform Wilcoxon signed-rank test on paired samples $x_{1,i}$ and $x_{2,i}$. I am trying to understand the independence assumption of the test. My questions are: Which quantities must ...
  • 486
2 votes
1 answer
1k views

When are time series data exchangeable?

E.g. is a linear process exchangeable? What about strict vs weak stationarity? EDIT: for clarity, I'm asking if/when the sequence of data points $(x_1, x_2, \ldots, x_n)$ in the time series is ...
  • 165
1 vote
1 answer
266 views

Exchangeability of group effects

Regarding this basic model with an individual-level covariate $x_{ij}$, group effects $U_{0j}$, and individual effects $R_{ij}$: $Y_{ij}=\gamma_{00}+\gamma_{10}x_{ij}+U_{0j}+R_{ij}$ the multilevel ...
  • 1,017
3 votes
1 answer
151 views

Create an exchangeable sequence from a non-exchangeable sequence

Suppose you have an arbitrary sequence of real values $\{ a_i | i \in \mathbb{N} \}$. Now, suppose you want to randomise the order of this sequence so that it is now exchangeable. To do this, you ...
  • 107k
11 votes
3 answers
4k views

Exchangeability and IID random variables

Every IID sequence of random variables is considered to be exchangeable, i understand why its necessary for the random variables to be identically distributed to assume exchangeability, but why the ...
  • 321
2 votes
1 answer
444 views

Understanding exchangeability

Suppose you had information of a true success rate of passing an exam from different schools, and that the 'true' success rates are similar, can you assume exchangeability is applicable to these ...
  • 321
3 votes
1 answer
86 views

Exchangeability and data smoothing

If a non-i.i.d sequence of a continuous random variable that is exchangeable, is smoothed by taking rolling average, is the resulting sequence exchangeable? My intuition suggests that it is not. I'll ...
  • 486
3 votes
1 answer
50 views

What role does stochastic independence have for Bayesians

I recently read a chapter about Subjectivist notions of Probability in Gillies (2012) and I stumbled upon the concept of exchangeability as the Bayesian version of independence (or something similar ...
  • 2,781
7 votes
2 answers
537 views

In a Bayesian hierarchical model, if exchangeability doesn't hold, what exactly goes wrong?

In many textbooks, when a Bayesian model is presented, such as a classic Normal-Normal model, there is some sort of brief mention that the trials must be exchangeable. I am wondering why this is ...
  • 2,365
2 votes
1 answer
64 views

Latent Modeling of Non-exchangable data or de Finneti on time-series

De Finneti's theorem applies on infinetely exchangeable sequences and allows building probabilistic models with latent variables that allow us to describe them. Is there a similar theorem for time-...
  • 924
1 vote
1 answer
304 views

Covariance of an exchangeable random vector

I am trying to solve the next exercise but I don´t know how to start. Let $X\in\mathbb{R}^n$ be a random vector which is exchangeable in the sense that $X$ has the same distribution as $(X_{\pi(i)})_{...
  • 21
3 votes
0 answers
321 views

Expected value of maxima of dependent random variables

I don't know if such theorem exists, but what I am looking for is a closed-form solution for $$E[\max(X_1, ..., X_N)]$$ where $X_1, ..., X_N$ is a sequence of dependent identically distributed ...