In Bayesian statistics a prior distribution formalizes information or knowledge (often subjective), available before a sample is seen, in the form of a probability distribution. A distribution with large spread is used when little is known about the parameter(s), while a more narrow prior ...

learn more… | top users | synonyms

2
votes
1answer
37 views

Forming a prior based on the solution to a linear system

On p. 115 of the 4th edition of Machine Learning a Probabilistic Perspective, we have the following: Let $\epsilon\sim N(0,\frac{1}{\lambda}\text{I})$ and let $L$ be a matrix of dimension ...
0
votes
0answers
10 views

Noninformative prior for variance, understanding and coding

I have three questions regarding the understanding behind and implementation of a noninformative prior for variance. I'm attempting to build a Metropolis sampler and I'm trying to sample from a ...
2
votes
0answers
24 views

Using empirical priors in PyMC

I'm using PyMC to sample the posterior distribution and I've run into a roadblock with using priors from samples, not models. My situation is as follows: I have some empirical data for a parameter ...
1
vote
2answers
284 views

Calculating SD for normal distribution with only mean and 5% and 95% quantile values

As part of a Bayesian method to estimate the divergence times of species, priors have to be set with values based on previous literature or known fossil dates. These priors can have different ...
2
votes
1answer
43 views

A simple question about MAP and MLE

I recently got this simple question from a friend. But I am quite confused about it. Suppose we toss a coin $N$ times, and got heads $m$ times. Assume the binomial distribution with $p$ which is the ...
2
votes
0answers
27 views

AR(1) model - which prior to use?

I want to use the following univariate model: $y_t = \mu_t + \epsilon_t, \ \epsilon_t \sim N(0,1)$ $\mu_t = \phi \mu_{t-1} + \omega_t, \ \omega_t \sim N(0,\sigma_\omega^2)$ That is, $\mu_t$ follows ...
0
votes
0answers
12 views

Variational Bayes with non-symmetric priors

I recently developed a Variational Bayes (VB) model with the intention encoding priors on a latent variable that was previously estimated using EM. In particular the VB model has two latent variables, ...
0
votes
0answers
22 views

A linear model with prior information

Suppose I have this experimental data: I have measurements of drug response from patients (let's say its blood pressure). Specifically, I have measurements after being treated with drug A (30 ...
0
votes
1answer
66 views

Equal weight between prior probabilities

While constructing a model hierarchy for Bayesian analysis, I have two parameters: $\theta_0$ ~ Uniform(80, 90) $\theta_1$ ~ Normal(0.093, 0.002) I take the $ln$ of the pdf for the parameter's ...
13
votes
2answers
471 views

What is the problem with empirical priors?

In literature I sometimes stumple upon the remark, that choosing priors that depend on the data itself (for example Zellners g-prior) can be criticized from a theoretical point of view. Where exactly ...
0
votes
1answer
44 views

Define own noninformative prior in stan

In the simple case of normally distributed data with unknown mean and variance, Jeffrey's prior is given by $$p(\mu, \sigma^2)=\frac{1}{\sigma^2}.$$ How can I define such a prior in the Stan ...
4
votes
2answers
104 views

Jeffreys Prior for normal distribution with unknown mean and variance

I am reading up on prior distributions and I calculated Jeffreys prior for a sample of normally distributed random variables with unknown mean and unknown variance. According to my calculations, the ...
0
votes
0answers
12 views

Issue in hierarchical model specification

I have an issue while trying to specify a model in a Bayesian framework for the following problem: With the goal of increasing statistical power when detecting associations between an outcome $y$ and ...
2
votes
1answer
21 views

Selecting correct priors for Bayesian A/B Testing

I'm running an A/B test. The test is a funnel that, at some step in the funnel, sends half the population to experience A, the other half to experience B. Traditionally everyone saw experience A at ...
0
votes
0answers
18 views

Set prior for logistic regression in R when using unequal group sizes (29 versus 48 cases)

I have 29 cases for negative outcome (0) and 48 cases for positive outcome (1). I fit my data with logistic regression model ...
1
vote
0answers
66 views

Relationship between low identifiability and prior weight in Bayesian model

I'm trying to get intuition into the relationship between low identifiability and prior weight in Bayesian model. Is it true to say that in lowly identifiable model + data the prior will have a higher ...
0
votes
0answers
9 views

Alpha Parameter Specification Dirichlet Prior

I have a straightforward Dirichlet-Multinomial model with code that is running in RJAGS. The data are a collection of 200 2 x 2 contingency tables. The multinomial counts are those of a 2 x 2 ...
0
votes
0answers
42 views

How to exploit relationships between independent variables?

Data: Each instance (representing a document) is a bag-of-entities (like BOW, except they're Wikipedia entities instead of words), so each feature is a binary or tfidf-like score based upon the ...
0
votes
0answers
27 views

Meaning of the prior and loss parameters in rpart in R

Could someone please explain to me what specifying priors and/or loss parameters in R's rpart actually do? I found R's documentation completely unhelpful. For example, let's suppose I have a highly ...
1
vote
1answer
87 views

Interpreting prior and posterior

I am bit puzzled on how we can interpret the posterior. Assume a coin which is 0.1 probable to be unfair. So our prior probability on the coin being unfair is 0.1, and being fair is 0.9. Also by ...
1
vote
0answers
37 views

Stochastic Block Model Priors

In the generic stochastic block model (binary edge data, no degree correction, etc.), if an uninformed prior is used for the Bernoulli coefficients i.e. Beta with $(a,b) = (1,1)$, will the model ...
-2
votes
2answers
1k views

Priors in Bayesian MCMC

I am trying to understand how the choice of priors affects a Bayesian model estimated using MCMC. At a basic level I understand that the product of the prior and the likelihood are proportional to the ...
0
votes
0answers
24 views

Jeffrey's Prior for normal distribution with mean = 0

How would I go about calculating Jeffrey's Prior for a normal distribution with mean = 0, So far I get: But then don't know where to go next. Any help much appreciated
7
votes
2answers
205 views

How does one use Bayes theorem with a continuous prior?

If my prior is modelled as a continuous probability distribution, say, a beta distribution skewed to reflect my bias towards certain models, how can I calculate the posterior probability? The ...
0
votes
0answers
23 views

Likelihood of hypothesis in live data

Bayes rule is $P(H|E)=\frac{P(H)P(E|H)}{P(E)}$ I have a prior distribution from categorical data prior={'a':0.2,'b':0.6,'c':0.1,'d':0.1} Which forms my ...
2
votes
2answers
63 views

Marginal likelihood vs. prior predictive probability

In the Bayesian framework, to me, it seems that the marginal likelihood and the prior predictive distribution/probability are equal. Is that the case? Or maybe this just holds for single data points? ...
0
votes
1answer
70 views

Why do we use Gamma($\epsilon, \epsilon$) as non-informative prior for precision and Normal prior for betas in Linear Regression

Suppose my regression model is $$Y_i = \beta_0 + \beta_1X_{i1} + \epsilon_i $$ In most books I am seeing that the prior used for precision $\tau = 1/\sigma^2 $ is $Gamma(\epsilon, \epsilon)$. However ...
0
votes
2answers
41 views

Return value of uniform distributions for MCMC simulations

I am confused about how what value should be returned from a uniform distribution when using MCMC simulations. The proper normal distribution is define as $$ p(\theta) = \left\{ \begin{array}{cc} ...
1
vote
1answer
38 views

Estimating von Mises Parameters for Angular Data

I want to model some angular data. Any input on how to incorporate the von Mises distribution and suggestions on appropriate priors in RJAGS for von Mises mean and concentration would be greatly ...
1
vote
2answers
46 views

Augmenting Kalman filter with parameter — what does the initial value mean?

It is a fairly standard trick to augment a Kalman Filter with unknown parameters and to propagate them forth with zero error to estimate them. I was wondering if anyone could tell me what the ...
3
votes
0answers
65 views

How to use information about likelihood of classes in a classifier?

General question: How can information about the likelihood of classes be used to improve a classifier? Suppose that the probability of each class is known quite precisely (from a very large sample), ...
4
votes
1answer
249 views

Is a spike-and-slab prior a proper prior?

Is a spike and slab prior a proper prior? (I am talking about a (product of Bernoulli) spike and Normal slab) If not, does it still lead to a proper posterior?
0
votes
0answers
22 views

Posterior predicted distribution, practical question

I'm new here to this place but I have already learned so much here. Yet I still remain with a large question involving my thesis in econometrics and medical scoence. For a starter, I have read ...
0
votes
1answer
33 views

Why using vague (or noninformative) priors? [duplicate]

In my Bayesian class, we are always required to specify vague (or noninformative) priors for bayesian modeling. I am quite confused about that. If I understand correctly, the main advantage of ...
0
votes
0answers
33 views

Variance of multinomial distribution that is product of 4 Beta random variables

I have a system of 4 binary random variables, $A$, $B$, $C$ and $D$. $A$, $B$ and $C$ are conditionally independent given $D$, and I'll call one set of samples $ABCD$ an event (e.g. $ABCD$ meaning all ...
2
votes
2answers
103 views

Can my Bayesian prior reflect what the data should say rather than what it could say?

Can my Bayesian prior reflect what the data should say rather than what it could say? For example, assume I collect data where $Y_i$ is whether or not student $i$ passed the test and $X_i$ is whether ...
1
vote
1answer
23 views

Prior for gamma distribution in “mean form”

I need to specify priors for the parameters of a gamma distribution. Normally the gamma distribution is parametrized in either the "rate-form'': ...
2
votes
0answers
22 views

Do mildly informative prior distributions tend to mitigate false positives (i.e. Type I error rates)?

I am curious if others have sources that speak to the matter that providing informative and/or mildly informative prior distributions on a parameter tend to mitigate false alarm rates? I know from the ...
2
votes
0answers
33 views

Incorporating population priors into MLE fits with few/limited samples

I am fitting Beta distributions to data resulting from each of many experiments using maximum likelihood. My goal is for each experiment, given iid data $y_{1:k}$, fit a Beta distribution, and then ...
1
vote
1answer
143 views

How to select hyperprior distribution for Beta distribution parameter?

I have a parameter $\theta$ whose value should lie between $(0,1)$. Therefore, I am assuming the prior distribution of $\theta$ to be a beta distribution with hyper-priors $\alpha$ and $\beta$ ie. ...
2
votes
0answers
24 views

prior for integer-valued random variable taking values 1 or greater

In my model I have an integer-valued random variable which should only take values one or greater. I would like to specify an appropriate prior for this which has most of the mass say around 1 to 5 ...
3
votes
1answer
46 views

Is it possible to define the mean of a varying distribution?

Suppose $(p_1,\ldots,p_k)$ be the vector of multinomial parameters and $$(p_1,\ldots,p_k)\sim \mbox{Dirichlet}(\alpha_1,\ldots,\alpha_k).$$ Let's define a function $f(p_1,\ldots,p_k) \in \mathbb{R}$. ...
1
vote
1answer
53 views

How is data generated in the Bayesian framework and what is the nature on the parameter that generates the data?

I was trying to re-learn Bayesian statistics (every time I thought I finally got it, something else pops out that I didn't consider earlier....) but it wasn't clear (to me) what the data generation ...
0
votes
0answers
27 views

Why does Empirical Bayes work in my simple case?

I have a problem where I am trying to classify data into two groups using a single parameter. The distribution of this parameter is Gaussian for two groups, so what I'm dealing with is two overlapping ...
1
vote
0answers
14 views

Creating a model for a webshop

I'm going to create a Multi-armed bandit algorithm to handle recommendations for a large scale webshop. I'm going to use Thompson sampling (http://en.wikipedia.org/wiki/Thompson_sampling) and would ...
3
votes
0answers
60 views

Prior for the coefficients of a linear regression model

I have a linear regression model $\bf Y=\bf{X}\bf{\beta}+\epsilon$. I want to assign a prior on $\bf\beta$ in order to derive the posterior predictive model $p(y_{predictive}|\bf{y},\bf{X},\beta)$. ...
2
votes
1answer
27 views

Iteratively solving for prior probabilites.

I'm using Bayes theorem to classify data into two groups, where the conditional probability is known but the prior is not. So I assume that the ratio of prior probabilities is 1 and calculate the ...
0
votes
0answers
14 views

How can i get prior information using my few data set from the whole data? [duplicate]

I have a data set (x1...x500, y1....y500 ) I want to know about bayesian regression I want to know the prior information , few data set(400) from the whole data (500) using MCMCregress( packages in ...
3
votes
1answer
23 views

Is this notation for the improper uniform prior correct?

Can I write: $\mu \sim U(0,\infty)$ ? Or do I have to use the notation $p(\mu) \propto 1$? Thank you.
1
vote
0answers
24 views

Can improper priors be implemented in some way?

I'm new to bayesian inference. I've just discovered that improper priors can't be specified in WinBUGS/OpenBUGS. I was wondering if this is common or not in bayesian inference. Are there same cases in ...