Questions tagged [monte-carlo]
Using (pseudo-)random numbers and the Law of Large Numbers to simulate the random behavior of a real system.
444
questions with no upvoted or accepted answers
9
votes
0
answers
151
views
In sports modelling, are hot simulations better or cold simulations?
I'm thinking here largely of the context in which someone has an Elo rating model for a particular sport.
To calculate things such as how often the team makes the Finals series, or wins the ...
9
votes
0
answers
2k
views
Expected value of softmax transformation of Gaussian random vector
Let $\mathbf w_1,\mathbf w_2,\ldots,\mathbf w_n \in \mathbb R^p$ and $\mathbf v \in \mathbb R^n$ be fixed vectors, and $\mathbf x \sim \mathcal N_p(\boldsymbol{\mu}, \mathbf{\Sigma})$ be an $p$-...
- 478
8
votes
0
answers
1k
views
Population Monte Carlo Algorithm
I am trying to wrap my head around the Population Monte Carlo Algorithm. I want to implement it for a mixture model, but I am uncertain on how to proceed. I am mostly looking for references or ...
- 3,641
7
votes
0
answers
89
views
Keeping track of the variance of a Metropolis-Hastings estimator
Let $(E,\mathcal E,\lambda)$ and $(E',\mathcal E',\lambda')$ be measure spaces, $p,q$ be probability densities on $(E,\mathcal E,\lambda)$, and $\varphi:E'\to E$ be bijective and $(\mathcal E',\...
- 419
7
votes
1
answer
84
views
Expected value of a "logistic uniform" multivariate
Let $\mathbf{a}_1,\ldots,\mathbf{a}_n \in \mathbb R^d$ and $b_1,\ldots,b_n \in \mathbb R$ be fixed. For $\mathbf{x} \sim \mathcal U([0,1]^d)$ and $j \in \{1,\ldots,n\}$, consider the real variable ...
- 478
7
votes
0
answers
746
views
Endogeneity in spatially lagged regression model
The standard convention in Spatial Statistic is that the spatial lag term in a regression model will be biased due to simultaneity. Looking at the following model, it would be difficult to argue with ...
- 71
6
votes
0
answers
789
views
Entropy of a mixture of Gaussians
I need to estimate as fast and accurately as possible the differential entropy of a mixture of $K$ multivariate Gaussians:
$$
\mathcal{H}[q] = -\sum_{k=1}^K w_k \int q_k(\textbf{x}) \log \left[\sum_{...
- 5,086
5
votes
1
answer
129
views
Looking for a recursive formula for asymptotic variance of importance sampling estimator (self-normalized)
Looking for a recursive formula to approximate variance of importance sampling estimator $Var_q\big[\delta_{IS}\big]\approx\sum_{i=1}^n\tilde w(X_i)^2\big[h(X_i)-\delta_{IS}\big]^2$. This is an ...
- 535
5
votes
0
answers
216
views
Sampling from mixture of *unnormalized* densities
Suppose I have $n$ unnormalized densities $g_1(\textbf{x}), \ldots, g_n (\textbf{x})$, for $\textbf{x} \in \mathbb{R}^d$, and $n \gg 1$, which largely overlap but in a nontrivial way. I need to sample ...
- 5,086
5
votes
0
answers
231
views
Computational entropy and Monte Carlo simulation
Is there a point at which the statistical properties of the random number generator will start to influence the results of Monte Carlo simulation?
I have a scenario where I need to calculate the ...
- 293
5
votes
0
answers
289
views
MCMC for Maximum Entropy?
Is there a way to sample from a discrete probability distribution, whose distribution itself is the solution to a Maximum Entropy problem with known linear constraints, without needing to solve for ...
- 1,487
5
votes
0
answers
191
views
Estimating parameters using Kullback-Leibler or Kolmogorov-Smirnoff via Nelder-Mead
I want to find the parameters of a model which specifies a set of classification probabilities, for say M classes. (I'll use the parameters in another model later.)
Given a set of parameters $\theta$,...
- 379
5
votes
0
answers
332
views
Deriving priors for MCMC implementation
I have been working on an assignment lately wherein the object is to implement an MCMC approach to simulate from a generated posterior distribution.
The posterior distribution is generated from a ...
- 1,521
4
votes
1
answer
81
views
Rao-Blackwellization in Black Box VI
In the paper, "Black Box Variational Inference," by Ranganath et al. (2013), the authors derive a Rao-Blackwellized estimator of the gradient of the evidence lower bound with respect to a ...
- 41
4
votes
0
answers
197
views
Why is my quasibinomial GLM estimator biased - Monte Carlo simulation
I'm playing with some Monte Carlo simulations to get an idea of the properties of some linear and non-linear models.
The linear OLS model in my case is specified as:
$Y_t = \beta_0 + \beta_1x+ \...
- 93
4
votes
0
answers
209
views
Any known approximations of summing quantiles from joint (bernoulli / lognormal) distributions
This is my first post to this site!
For an insurance-like scenario, I have several independent risks which I want to sum together and find a 95% percentile. Currently I do this by Monte Carlo but I ...
- 201
4
votes
0
answers
634
views
Deriving the bias of the self-normalizing importance sampling estimator
Suppose we are interested in the expectation of a test function $f(X)$ with respect to target distribution $\pi(X) \propto \gamma(X)$ using importance sampling with proposal distribution $q(X)$ with $...
- 2,204
4
votes
0
answers
2k
views
Detailed Balance for Hamiltonian Monte Carlo
I am trying to understand the detailed balance proof present in this paper: https://arxiv.org/abs/hep-lat/9208011v2 (page 4).
My question: Why do we consider the volume of a neighborhood of points ...
- 141
4
votes
1
answer
2k
views
What is the best way to report the results and uncertainty from a Monte Carlo simulation?
I am fitting data to a model that has ~30 input parameters, each with their own uncertainty levels, and which can interact with each other in the model. I therefore decided the best way to fit the ...
- 141
4
votes
0
answers
765
views
Modeling a Correlated Bivariate Beta Distributions in PyMC3
My goal is to perform a bayesian A/B test of probabilities of success in two groups considering a hypothesis about non-zero covariance between those probabilities.
Bivariate beta distribution
I am ...
4
votes
0
answers
667
views
Convergence Proof of First Visit Monte Carlo Control
I am currently trying to find a formal proof of convergence for the Monte Carlo Reinforcement Learning Methods described in Sutton,Barto's Book "Reinforcement Learning - An Introduction" , Section 5.
...
- 193
4
votes
0
answers
415
views
Convergence of gradient descent Monte Carlo Control with function approximation
Can anyone point me in the direction of a formal proof of convergence for a (on/off policy) Monte Carlo control algorithm with (non-)linear function approximation?
In http://incompleteideas.net/book/...
- 141
4
votes
0
answers
81
views
Sampling from distribution defined variationally
Suppose I define a probability distribution $\mu\in\mathrm{Prob}(\Sigma)$ over some compact $\Sigma\subseteq\mathbb{R}^n$ using a variational problem:
$$
\mu:=\arg\min_{\mu\in\mathrm{Prob}(\Sigma)} F[\...
- 799
4
votes
0
answers
947
views
How to Validate a Monte Carlo Simulation
I have historical data of a production process, and I've being asked to build a simulation model to predict its performance in the future. Using the historical data, I've being able to obtain the ...
4
votes
0
answers
104
views
Uniform convergence of Monte Carlo approximation
Usually Monte Carlo method is used to compute integration. For example, let $g(x,\theta)$ be a continuous function about $x$ and $\theta$, $f(x \mid \theta)$ is a continuous pdf with parameter $\theta$...
- 41
4
votes
0
answers
229
views
How to show that the variance of Sequential Importance Sampling estimates increase with the dimension?
I am trying to understand the Particle Filter and the motivation to use it over the regular Sequential Importance Sampling. As far as I understand until now:
1- We try to estimate the expectation of ...
- 2,148
4
votes
0
answers
163
views
How to estimate the given function using Rao Blackwellization approach?
I have a function $X$ which is lognormal (0,1)
and then another function
$\log Y = 4 + 2 \log(X) + \epsilon$
where $\epsilon \sim \mathcal N(0,1)$
I want to estimate $E(Y|X)$ as a Rao Blackwellized ...
- 41
4
votes
0
answers
345
views
Gibbs sampling from full conditionals
I have the following joint density:
$p(x_1,x_2,y_1,y_2) \propto \exp\left(−\left(x_1^2+x_2^2+c_1(y_2-y_1)^2+c_2(y_2-y_1)^4\right)\right)$
Can I use Gibbs sampling to sample from that? How can I get ...
- 41
4
votes
0
answers
72
views
Random walk with restricted graph knowledge
I have a very large graph and a function of its vertices, and want to estimate mean value of this function.
It's not possible to sample vertices uniformly in this problem, so a reasonable choice for ...
- 63
4
votes
0
answers
112
views
Averaged estimators in stochastic versions of EM
Recently I've been working EM algorithms for MAP estimation in a problem where the expectation is intractable, but the maximization is easy. Further, draws from the distribution in the E-step are ...
- 4,750
4
votes
1
answer
265
views
Is there an intuitive way to understand a sequential Monte Carlo Markov Chain?
I was wondering if there was an intuitive way to understand what a sequential Monte Carlo Markov Chain is. It also goes by the name of particle filtering. Is there an intuitive way to think about it/...
- 2,057
3
votes
0
answers
242
views
How to calculate the uncertainty of a parameter when using a cost function other than the least squares?
Context:
Given some measurements $y_i$ associated with the independent variable $x$ and uncertainties $\sigma_i$ and, on the other hand, a model $f(x;\theta)$ where $\theta$ is a free parameter (which ...
- 31
3
votes
0
answers
134
views
A generalized randomized mean estimate based on the Chebyshev inequality
Let $X$ be a random variable that does not take on a single value with probability 1. Let “black-box" sample access to the random variable $X$ be given. Let $M_k$ be an upper bound on the $k$th ...
- 1,144
3
votes
0
answers
40
views
Computationally + Statistically Efficient Unbiased Estimation of Chebyshev Polynomials of Expectations
Let $T_n$ denote the $n^\text{th}$ Chebyshev polynomial, defined by the recursion
\begin{align}
T_0(x) &= 1,\\
T_1(x) &= x,\\
T_n(x) &= 2x \cdot T_{n-1} (x) - T_{n-2} (x).
\end{align}
Now, ...
- 1,316
3
votes
0
answers
124
views
How are the variance of an estimate to $\int_Bf\:{\rm }d\mu$ and the deviation of $f$ from the mean $\frac1{\mu(B)}\int_Bf\:{\rm d}\mu$ related?
Let $(E,\mathcal E,\mu)$ be a probability space, $(X_n)_{n\in\mathbb N_0}$ be an $(E,\mathcal E)$-valued ergodic time-homogeneous Markov chain with stationary distribution $\mu$ and $$A_nf:=\frac1n\...
- 419
3
votes
1
answer
105
views
Erroneous expression for Metropolis-Hastings acceptance ratio in a paper
Let
$(E,\mathcal E)$ be a measure space;
$\rho:E\to[0,\infty)$ be $\mathcal E$-measurable, $p:E^2\to[0,\infty)$ be $\mathcal E^{\otimes2}$-measurable, $$r(x,y):=\left.\begin{cases}\displaystyle\frac{\...
- 419
3
votes
0
answers
46
views
Probability that one Gaussian RV exceeds all others
Imagine we have $k$ Gaussian RVs $$
X_i \sim N(\mu_i, \sigma_i^2) \text{ for } i=1, \ldots, k
$$
and we sample from each of them independently to produce a vector, $\vec{x} = (x_1, \ldots, x_k)$.
For ...
- 611
3
votes
0
answers
300
views
Why do we use the log-derivative trick before Monte Carlo?
I still don't understand how we can approximate the gradient of an expected value... Indeed it's impossible to sample points and then to average the gradients of them as we have only samples... (How ...
- 387
3
votes
1
answer
84
views
Numerical examples proving and disproving the optimal scaling heuristic by Roberts et al
Let
$d\in\mathbb N$ with $d>1$
$\ell>0$
$\sigma_d^2:=\frac{\ell^2}{d-1}$
$f\in C^2(\mathbb R)$ be positive with $$\int f(x)\:{\rm d}x=1$$ and $g:=\ln f$
$Q_d$ be a Markov kernel on $(\mathbb R^...
- 419
3
votes
0
answers
19
views
Visualization of unbiasedness of high dimensional paramter estimates
Assume a statistical model $f_{\theta}(X)$ that allows to estimate a parameter vector $\hat{\theta}\in \mathbb{R}^p$ from data $X$ and assume that $p$ is high dimensional (you may assume something ...
- 1,438
3
votes
0
answers
26
views
References for Texas Holdem
Does anybody know any good papers or software that use Monte Carlo techniques to estimate the probability of certain hands or winning/losing a hand in Texas Holdem?
Ideally I'd like to have some ...
- 19.4k
3
votes
0
answers
106
views
Hamiltonian Monte-Carlo in Reinforcement Learning
I was wondering if someone can point me out a reinforcement learning algorithm/paper (continuous variable), in which the Hamiltonian Monte-Carlo has been directly used. Thanks!
- 131
3
votes
1
answer
213
views
Gibbs Sampling vs. Using Raw Probability in Contrastive Divergence
In Hinton's Practical Guide to Training Restricted Boltzmann Machines, Section 3, he discusses different situations in which one should take a sample from the Gibbs sampling process, and other ...
- 61
3
votes
0
answers
829
views
Estimating standard error of Monte Carlo integration, non-MCMC version
Let us suppose that we're to evaluate the expectation of a random variable $h$ with respect to some distribution $\pi$, $\text{E}_{\pi}[h]$. The standard Monte Carlo estimate, using a sample of $X_1, ...
- 131
3
votes
0
answers
184
views
Monte carlo method and Convergence in Distribution
Monte Carlo method, from what I could gather, allows one to obtain observations/draws from a possibly unknown statistical distribution. Let's say $T(X)\sim F$, where $T$ is a statistic, $X$ is a ...
- 5,330
3
votes
0
answers
629
views
Importance sampling: what is this bias?
I am experiencing what seems to be a bias in importance sampling, which, given that it's an unbiased procedure, should not be there.
Consider linear regression
$$
y = X\beta+\epsilon
$$
where there ...
- 2,368
3
votes
0
answers
179
views
Error bars for sum of Monte Carlo integrations
I want to calculate a quantity $I$ which is the sum of integrals $I_k$,
$$
I = \sum_k I_k\;,\quad I_k = \int_{\Omega} f_k(x)\,\textrm d x\;,
$$
Every integral $I_k$ is evaluated independently by a ...
- 251
3
votes
0
answers
133
views
sampling from a posterior derived from hierarchical Bayesian using HMC
I have a complex pdf based on hierarchical Bayesian formalism where x depends on the priors w'and w'', and I consider hyper-prior for the latter's as w=Php(zeta,beta) where Php stands for the hyper ...
- 55
3
votes
0
answers
67
views
Confusion on three “types” of Markov Chain Monte Carlo for the same inference
This is a long question but I would be very grateful if someone can help or provide some reference! And I believe this is a common confusion among beginners of MCMC.
Background
Given a directed graph $...
- 1,763
3
votes
0
answers
1k
views
Monte Carlo simulation for fitting distributions (Weibull and log-normal)
I am working on a computer project which needs statistical analysis and I am not much of a statistic person. I have a Bluetooth (BT) detector device which detects passing Bluetooth devices (i.e one ...
- 193