The reversible-jump Markov chain Monte Carlo state comprises the model index and that model's parameter. Its unnormalized posterior density in model is
For models of equal dimension, choose model with probability and propose using density on the destination parameter space. The equal-dimension reversible-jump acceptance probability is
On acceptance change both model and parameter; on rejection retain both. Choose model proposals connecting all models of positive posterior mass and within-model kernels exploring their supports, with aperiodicity, to obtain an ergodic chain. Include within-model updates targeting its conditional posterior, and use model occupation proportions after the initial transient to estimate posterior model probabilities. The formula includes the model prior probabilities, parameter prior densities, likelihoods, reverse model-selection probability, and reverse parameter-proposal density.
This direct-density formula is Metropolis–Hastings algorithm on a disjoint union of equal-dimensional spaces. If instead a move uses a deterministic bijection with auxiliary proposal densities and , use
The Jacobian determinant is one for identity matching, but equal model dimensions alone do not make a nonlinear map volume-preserving. For a direct proposal density the change of variables is already included in that density, so no additional Jacobian factor is inserted.
For the two Poisson models, use the shape-rate convention for the gamma distribution:
If denotes scale instead, replace every occurrence of the rate below by . Introduce the inactive under model with the proper pseudo-prior , independent of . This pseudo-prior augmentation for model comparison does not change model 's marginal likelihood because the inactive density integrates to one. The two augmented targets on the same positive quadrant are
where
Use a symmetric cross-model proposal that flips the model label and leaves both parameters unchanged. The matching map is the identity, with unit Jacobian determinant, and the common prior factors cancel. Hence the model-switch acceptance probabilities are
Only the second observation's factor changes because is retained as its shared candidate mean. Evaluate the ratio on a logarithmic scale as for the first direction.
For within-model moves, Poisson-gamma conjugacy gives exact Gibbs sampling refreshes:
The two draws are conditionally independent within either model. Choose with fixed positive probabilities between this block refresh and the cross-model proposal. Both moves preserve the augmented posterior; refreshing the inactive parameter maintains its correct distribution and supports mixing. The fraction of retained states labelled estimates .
An equivalent literal dimension-changing implementation deletes on a proposed move, and on a move draws an auxiliary and sets . The birth matching has dimensions and unit Jacobian determinant. Its proposal density cancels the new parameter's prior density, giving the same acceptance ratios above for symmetric selection of move directions. This provides suitable moves without storing an inactive parameter.
For an independent check of Poisson mean equality model comparison, the exact Bayes factor is
It follows by integrating the two gamma likelihood kernels; the common factorial terms cancel. With equal model priors, the exact posterior probability of is , which can also be used to check the RJ-MCMC occupation estimate.
For the Box-Muller transform, draw independent from the uniform distribution on and set
The radial density is for , with an independent uniform angle. The polar-to-Cartesian Jacobian determinant is , so the joint density of is
This factorization proves that both outputs have the standard normal distribution and are independent. Repeating the Box-Muller transform with fresh independent uniform pairs gives independent normal outputs; discard one extra output if the desired sample size is odd.
For the prescribed binary probabilities, generate independent standard normal values in this way and use
The standard normal distribution function gives , and independence is preserved because each threshold uses a different independent normal value.
For the probit regression posterior, introduce latent variables
Given , these are independent variables, and normal symmetry gives . Thus this data augmentation has exactly the observed binary likelihood. With the specified normal prior distribution, the augmented joint posterior is proportional to
The latent-normal Gibbs sampler for probit regression alternates two blocks. First, given the current , draw each independently from its truncated normal distribution, namely restricted to the sign fixed by . One exact method is to use the Box-Muller transform for , form , and reject until the sign is correct. The probability of success is positive at every finite parameter value, so the method is valid, although it can be slow for a rare sign.
For direct sign-truncated normal sampling, let , and draw . An inverse transform sampling implementation is
Use suitable tail or survival-function evaluations when floating-point probabilities approach zero or one; the rejection construction remains a valid alternative.
Second, completing the square in yields its full conditional distribution:
Generate a fresh standard normal value by the Box-Muller transform, multiply by the conditional standard deviation, and add the conditional mean. Start from any finite , alternate these steps, discard an initial transient and use the retained values to approximate its posterior distribution. These are dependent Markov chain Monte Carlo samples, rather than independent posterior draws. Each block is an exact Gibbs sampling update for the augmented posterior, whose marginal in is the requested posterior.
For a real weakly stationary process, extend its autocovariance by . The exact existence of a time-series spectral density condition is that its spectral measure of a stationary time series be absolutely continuous with respect to Lebesgue measure. Absolute summability is a useful sufficient condition, not a necessary one.
We use the conventional angular-frequency density on , restricted to by symmetry. Under the absolute-summability condition, the Fourier series and its inverse relation are
Thus . If a one-sided density is instead normalized to integrate to the full variance, use and omit the factor two in the inverse formula. This is the positive-frequency spectral normalization convention difference. With merely an integrable spectral density, the inverse relation remains valid; one must not assume pointwise convergence of the unweighted Fourier series. Its Fejér sums recover the density in :
If and are independent stationary processes, the cross covariances vanish, so
Linearity of the inverse Fourier series relation, or addition of the spectral measures of a stationary time series, gives . Independence can in fact be weakened to zero cross-covariances at every lag.
For the ARMA representation, away from an uncancelled unit root the transfer function gives
The usual causal case has ; the same expression holds for the two-sided stationary solution when . We work in the nondegenerate case and ; cancellations and zero-noise cases are obtained by the appropriate reduced representation or limits. Since independent white noise of variance contributes , put
Then
The white-noise addition to an ARMA(1,1) process problem is therefore the factorization
Define
Both are positive under the stated nondegeneracy conditions. An invertible choice is
These satisfy , , and . Hence
In particular, it is the autoregressive coefficient that enters the terms from the added observation noise.
To obtain a representation of the actual , define
The stable inverse exists because , and spectral filtering gives . Thus is weak white noise, and
Its variance can equivalently be written
When , also ; when , necessarily and , so the latter quotient should not be used. The model may reduce in order through cancellation. Boundary limits with or give and need not be stably invertible; the displayed stable reconstruction applies to the nondegenerate case above.
In the classical risk model write , where the Poisson process has rate and is independent of the claim sizes. Define the ruin time and ultimate ultimate ruin probability by
The relative safety loading is , so . The adjustment coefficient is the nonzero positive solution
Existence and uniqueness follow from the secant-slope existence criterion for an adjustment coefficient. Explicitly, has derivative at zero and is strictly convex for positive claims. The assumed divergence of makes it cross zero once at a positive . When , an exponential lower bound from any positive tail event shows that . In particular is inside the finite-transform domain, not at its endpoint.
Put . Replacing in the survival renewal equation for a classical risk model yields
By the tail integral formula for moments, . Therefore this is the defective renewal equation
The original kernel has mass . For the exponential tilt of the ruin renewal kernel, set
Multiplying by gives the required proper renewal equation
To verify that this is a probability renewal kernel, Tonelli theorem gives, for in the finite-transform domain,
The adjustment coefficient equation consequently implies . Its expected value is
It is finite because is interior to the finite-transform domain and is positive because the strict convex crossing has derivative .
We quote the key renewal theorem in the following form: for a nonarithmetic distribution of positive increments with finite positive mean , and a directly Riemann integrable nonnegative function , the locally bounded solution of satisfies . It has renewal representation ; this follows by iterating the equation, since the probability that arbitrarily many positive increments have sum at most a fixed tends to zero.
Here is absolutely continuous, and hence nonarithmetic distribution, even if the original claim law has atoms. The function is continuous. Choose with . The Markov inequality gives , and hence . Continuity on compact intervals and this exponential bound make the upper Riemann sums finite with uniformly vanishing tails, proving direct Riemann integrability.
A further application of Tonelli theorem evaluates the forcing integral:
All hypotheses of the key renewal theorem are now checked, so the interior adjustment coefficient ruin prefactor is
This proves the requested Cramér–Lundberg ruin asymptotic, including its constant.
For the two-component hyperexponential distribution, conditioning on the chosen exponential component gives
Its moment-generating function and derivative are
The moment-generating function diverges at , while the derivative of the adjustment equation at zero is negative. Strict convexity therefore places its unique positive adjustment coefficient in
The adjustment equation, divided by , becomes
Using this identity to subtract from gives
Thus the asymptotic constant in terms of and is
The denominator is positive on the identified domain. If desired, is the smaller root of ; the other algebraic root lies outside the positive finite-transform interval and is not an adjustment coefficient.

Pinned article: Introduction to the OurBigBook Project

Welcome to the OurBigBook Project! Our goal is to create the perfect publishing platform for STEM subjects, and get university-level students to write the best free STEM tutorials ever.
Everyone is welcome to create an account and play with the site: ourbigbook.com/go/register. We belive that students themselves can write amazing tutorials, but teachers are welcome too. You can write about anything you want, it doesn't have to be STEM or even educational. Silly test content is very welcome and you won't be penalized in any way. Just keep it legal!
We have two killer features:
  1. topics: topics group articles by different users with the same title, e.g. here is the topic for the "Fundamental Theorem of Calculus" ourbigbook.com/go/topic/fundamental-theorem-of-calculus
    Articles of different users are sorted by upvote within each article page. This feature is a bit like:
    • a Wikipedia where each user can have their own version of each article
    • a Q&A website like Stack Overflow, where multiple people can give their views on a given topic, and the best ones are sorted by upvote. Except you don't need to wait for someone to ask first, and any topic goes, no matter how narrow or broad
    This feature makes it possible for readers to find better explanations of any topic created by other writers. And it allows writers to create an explanation in a place that readers might actually find it.
    Figure 1.
    Screenshot of the "Derivative" topic page
    . View it live at: ourbigbook.com/go/topic/derivative
  2. local editing: you can store all your personal knowledge base content locally in a plaintext markup format that can be edited locally and published either:
    This way you can be sure that even if OurBigBook.com were to go down one day (which we have no plans to do as it is quite cheap to host!), your content will still be perfectly readable as a static site.
    Figure 5. . You can also edit articles on the Web editor without installing anything locally.
    Video 3.
    Edit locally and publish demo
    . Source. This shows editing OurBigBook Markup and publishing it using the Visual Studio Code extension.
  3. https://raw.githubusercontent.com/ourbigbook/ourbigbook-media/master/feature/x/hilbert-space-arrow.png
  4. Infinitely deep tables of contents:
    Figure 6.
    Dynamic article tree with infinitely deep table of contents
    .
    Descendant pages can also show up as toplevel e.g.: ourbigbook.com/cirosantilli/chordate-subclade
All our software is open source and hosted at: github.com/ourbigbook/ourbigbook
Further documentation can be found at: docs.ourbigbook.com
Feel free to reach our to us for any help or suggestions: docs.ourbigbook.com/#contact