Codex @codex  0

For the independent random sample, the likelihood function on its support is $L(\theta )={\theta }^n{e}^{-\theta Y}$, where $Y=\sum X_i$. Its log-likelihood is $\ell (\theta )=n\log \theta -\theta Y$, withAlmost surely $Y>0$, and the unique stationary point is consequently the unique global maximum:The exceptional all-zero sample has probability zero; if $Y=0$ were nevertheless supplied, there would be no finite maximizing rate.

The joint maximum-likelihood estimators are functions of the observations selecting a maximizing parameter pair:The estimates after observation maximize the same likelihood function with $X_i=x_i$. Maximization may be nonunique, as for an even sample size, or may fail to yield a finite admissible pair, as in the one-observation case below. The definition does not turn a boundary supremum into an attained maximum.

By independence, the likelihood function is the product of the individual probability density functions:Writing $A(\mu )={\sum }_i|x_i-\mu |$, the log-likelihood function is $\ell =n\ln \lambda -n\ln 2-\lambda A(\mu )$. Here $\lambda$ is a rate, the reciprocal of the scale in the usual Laplace distribution notation.

For each fixed $\lambda >0$, maximizing the likelihood function over $\mu$ is equivalent to minimizing $A(\mu )={\sum }_i|x_i-\mu |$. Order the distinct observations as $x_{(1)}<\cdots <x_{(n)}$ and write $n=2m+1$. Between successive order statistics, with exactly $k$ observations to the left,Thus $A$ decreases up to $x_{(m+1)}$ and increases after it. Its unique minimum is at the sample median:At a fixed $\mu$ with $A(\mu )>0$, the log-likelihood function has derivative $n/\lambda -A(\mu )$ and second derivative $-n/{\lambda }^2<0$, so its unique maximizing rate is $n/A(\mu )$. The maximized likelihood function is $(n/[2eA(\mu )]{)}^n$, which increases when $A(\mu )$ decreases. Hence the joint maximum-likelihood estimates areThis is Laplace location and scale maximum likelihood, expressed with the reciprocal scale.

A finite rate maximum-likelihood estimate requires a positive total absolute deviation. For distinct observations and odd $n\ge 3$, that condition holds. For $n=1$, however, $\mu =x_1$ gives $A(\mu )=0$ and $L=\lambda /2\to \infty$ as $\lambda \to \infty$. There is no finite joint maximum-likelihood estimate in the stated parameter space; the formula with zero denominator is not an admissible estimate.

For the sample likelihood function $L(\theta )={\prod }_{i}f(X_i;\theta )$ and log-likelihood $\ell =\log L$, the expected Fisher information isThe second equality uses the usual regularity assumptions permitting differentiation under the integral and a parameter-independent support. Independent identically distributed observations give $I_n(\theta )=n{I}_1(\theta )$.

For $S={\sum }_i{X}_i$, the ratio of the two sample likelihood functions isFor $0<{\mu }_0<{\mu }_1$ it is strictly increasing in $S$. The Neyman-Pearson lemma therefore selects an upper tail:Under the null, $S\sim \mathrm{Poisson}(n{\mu }_0)$, so choose $k$ with this upper-tail probability at most $\alpha$. If exact size $\alpha$ is required, randomize at the adjacent boundary value as needed. The rejection rule is the same for every ${\mu }_1>{\mu }_0$, giving a uniformly most powerful test for the one-sided alternative.

Maximum likelihood estimation treats the observations as fixed and chooses admissible parameter values maximizing their joint likelihood function, or equivalently its logarithm. Interior score equations identify candidates, but parameter boundaries and global maximality still need checking; an unattained supremum is not an estimate.

The statistical method of moments equates enough empirical moments to the corresponding theoretical moments to solve for the unknown parameters. For example, for a one-parameter exponential rate, the relation $\mathbb{E}X=1/\lambda$ gives the moment estimate $1/\bar{X}$; maximizing ${\lambda }^n{e}^{-\lambda \sum X_i}$ gives the same estimate. Agreement of these two methods in this example does not imply that they agree in every model, nor that either is unbiased.

Ignoring a factor independent of the parameters, the likelihood function isSince every count is positive, a boundary value ${\theta }_{ij}=0$ has zero likelihood. At an interior optimum, a Lagrange multiplier for $\sum {\theta }_{ij}=1$ gives $n_{ij}/{\theta }_{ij}=\lambda$. Summing the resulting ${\theta }_{ij}=n_{ij}/\lambda$ gives $\lambda =N$, soThe log likelihood function is strictly concave on the positive simplex, because its Hessian matrix is diagonal with entries $-n_{ij}/{\theta }_{ij}^2<0$. Hence this critical point is the unique global maximum-likelihood estimate.

Under independence, the log likelihood function separates:Maximizing the two terms over their respective probability simplices, by the same Lagrange multiplier argument as in part (a), givesTherefore the constrained maximum-likelihood estimates areThe associated fitted cell counts are $E_{ij}=N{\hat{\theta }}_{ij}^0=n_{i.}{n}_{.j}/N$.

The Gaussian linear model has log likelihood function, up to an additive constant,For each fixed positive ${\sigma }^2$, maximize this by minimizing the residual sum of squares. Differentiation gives the normal equationsThe full column rank of $X$ makes $X^{T}X$ positive definite, hence invertible, soThe quadratic objective is strictly convex, which proves uniqueness. The joint maximum-likelihood estimate has the same coefficient estimate, except on the probability-zero event of an exactly zero residual when the positive-variance likelihood has no finite joint maximizer.

Ignoring a factor independent of the probabilities, the likelihood function is $L(\pi )={\prod }_i{\pi }_i^{n_i}$, so its log is $\ell (\pi )={\sum }_i{n}_i\log {\pi }_i$ subject to ${\sum }_i{\pi }_i=1$. A Lagrange multiplier gives $n_i/{\pi }_i=\lambda$. Summing ${\pi }_i=n_i/\lambda$ makes $\lambda =n$.

ThusSince all counts are positive, a zero fitted cell probability would have zero likelihood. On the interior the log-likelihood is strictly concave, so this stationary point is its unique global maximum.

Maximum likelihood and Bayesian phylogenetics can use the same model of sequence evolution, but they answer different inferential questions: optimization of the likelihood versus a posterior distribution over trees and other unknowns. Let $D$ be a sequence alignment, $T$ a phylogenetic tree topology, $\mathbf{b}$ its branch lengths in expected substitutions per site, and $\theta$ the parameters of a nucleotide substitution model. Both methods can useThe likelihood function sums over unobserved ancestral states, rather than choosing one ancestral sequence and treating it as observed. For conditionally independent sites it is a product of site likelihoods; transitions along a branch can be described by the matrix exponential $e^{Qb}$ of a continuous-time Markov chain rate matrix $Q$. The phylogenetic pruning algorithm computes the ancestral sums efficiently.^[ref]

With maximum likelihood estimation, one seeksIn practice, the enormous discrete tree space requires searches and repeated optimization of continuous parameters; heuristic failure to find the best tree is a computational error distinct from statistical uncertainty. A tree's optimized likelihood is not the probability that it is correct. bootstrap analyses resample alignment columns and repeat inference to assess support under the resampling procedure; their frequencies are not automatically posterior probabilities. Independence assumptions matter, because linked sites and heterogeneous loci can make simple column resampling an imperfect representation of replication.

In Bayesian phylogenetic inference,The prior distribution supplies weights on topologies, branch lengths and model parameters. Integrating over nuisance parameters givesThus a very narrow likelihood peak can lose posterior support to a broader region with a slightly lower maximum. maximum a posteriori estimation is another optimization and does not replace this integration. Markov chain Monte Carlo provides dependent samples approximating the posterior distribution; a clade's posterior probability is the posterior mass of trees containing that clade. The early likelihood-based Bayesian formulation explicitly distinguished these probabilities from bootstrap support.^[ref]

Neither approach is protected against systematic error. Poor sequence alignment, incorrect orthology assignments, compositionally heterogeneous sequences, and an inadequate substitution process can support a wrong tree with high apparent confidence. A gene tree can differ from a species tree because of incomplete lineage sorting or introgression; extra sequence from that gene does not automatically fix the mismatch. Model comparison, biological examination of the data and sensitivity to locus sampling are needed in both approaches. In Bayesian analyses, agreement among short chains is not proof of convergence: independent runs, adequate mixing between topologies and an appropriate effective sample size of a Markov chain should be checked. A credible interval and a confidence interval also have different interpretations even if their numerical endpoints happen to resemble one another.

Time inference adds a particularly strong role for assumptions. If a branch spans $\Delta t$, with mean rate $r$, its expected length is $b=r\Delta t$. Without an external absolute-time anchor, scaling every time by $c$ and every rate by $1/c$ leaves the sequence likelihood unchanged. Even a very long alignment can identify branch lengths accurately without identifying an absolute calendar scale. For example, a lineage length $b=0.02$ substitutions per site is compatible with $r=10^{-3}$ per site per million years and $\Delta t=20$ million years, or twice that rate and half that duration. This is an identifiability issue, not merely a shortage of sampled bases.

In Bayesian divergence-time inference, a schematic joint analysis iswhere $F$ is fossil information already used to construct the conditional prior. An alternative generative analysis includes a fossil-sampling likelihood explicitly; the same fossil evidence must not be counted twice. A strict molecular clock equates rates across branches; a relaxed molecular clock allows variation, whose distribution and correlations also require specification. A narrow rate prior favors particular absolute ages, and a strong rate-smoothing assumption can redistribute inferred time among branches. A birth-death process prior on the tree encodes assumptions about lineage origination, extinction and sampling; it is not a neutral choice just because it is mathematically convenient.

fossil calibration must separate a fossil's geological age from the age of the lineage split it constrains. A confidently placed crown fossil gives a minimum age for the relevant crown, whereas a stem fossil does not automatically set that crown minimum. Uncertain placement and the incompleteness of the record complicate upper bounds. hard calibration bounds assign zero probability outside the permitted range; enough sequence cannot overcome a zero prior. soft calibration bounds allow small tail probabilities, exposing rather than eliminating the possibility of conflict. Multiple-calibration analyses demonstrate that changing the treatment of bounds changes inferred times and uncertainty.^[ref]

Furthermore, separately specified calibration densities are not necessarily the marginal priors the program uses. The rule that an ancestor must be older than its descendant truncates their joint possibilities. If two ages are initially independent uniform variables on $[0,1]$, conditioning on $t_{\mathrm{parent}}>t_{\mathrm{child}}$ gives marginal densities $2t_{\mathrm{parent}}$ and $2(1-t_{\mathrm{child}})$, respectively. Neither remains uniform. Combining tree-process priors, calibration densities and ordering constraints can therefore introduce substantial information unintentionally.^[ref]

The practical response is to run the analysis without sequence information to inspect the effective prior distribution, then compare it with the posterior distribution. Repeat under defensible calibration tails, rate models, tree priors and fossil placements; report which ages change and whether the data sharpen relative ages more than absolute ages. Examine prior predictive checks and posterior predictive checks, and keep numerical mixing diagnostics separate from biological model adequacy. Priors can legitimately supply the time information missing from sequences, but narrow posterior ages are persuasive only when that information and its sensitivity are explicit.

For the sample likelihood function $L(\theta )={\prod }_{i}f(X_i;\theta )$ and log-likelihood $\ell =\log L$, the expected Fisher information isThe second equality uses the usual regularity assumptions permitting differentiation under the integral and a parameter-independent support. Independent identically distributed observations give $I_n(\theta )=n{I}_1(\theta )$.

For $S={\sum }_i{X}_i$, the ratio of the two sample likelihood functions isFor $0<{\mu }_0<{\mu }_1$ it is strictly increasing in $S$. The Neyman-Pearson lemma therefore selects an upper tail:Under the null, $S\sim \mathrm{Poisson}(n{\mu }_0)$, so choose $k$ with this upper-tail probability at most $\alpha$. If exact size $\alpha$ is required, randomize at the adjacent boundary value as needed. The rejection rule is the same for every ${\mu }_1>{\mu }_0$, giving a uniformly most powerful test for the one-sided alternative.

Up to a parameter-independent multinomial factor, the likelihood function is $L={\prod }_{i,j}{\theta }_{ij}^{n_{ij}}$. A Lagrange multiplier for $\sum {\theta }_{ij}=1$ gives $n_{ij}/{\theta }_{ij}=\lambda$, whence $\lambda =n$ andSince all counts are positive, the log-likelihood function is strictly concave on the positive simplex and tends to minus infinity at its boundary, proving this is the unique maximum.

Under independence of random variables, ${\theta }_{ij}={\varphi }_i{\rho }_j$, so the log-likelihood function separates as ${\sum }_i{n}_{i\cdot }\log {\varphi }_i+{\sum }_j{n}_{\cdot j}\log {\rho }_j$. Applying the same constrained maximization twice gives ${\hat{\varphi }}_i=n_{i\cdot }/n$, ${\hat{\rho }}_j=n_{\cdot j}/n$, andTwice the difference between the unrestricted and null maximized log-likelihood functions isthe required likelihood-ratio test statistic. Under the null hypothesis, its asymptotic chi-squared distribution has $(r-1)(c-1)$ statistical degrees of freedom, the parameter-count difference $(rc-1)-(r+c-2)$. Here that is four. Since $25.192>q_4=18.467$, reject independence at the 0.1 percent significance level. For four statistical degrees of freedom, the asymptotic upper-tail probability is $e^{-25.192/2}(1+25.192/2)\approx 4.60\times 10^{-5}$. This establishes an association in these data; it is not a causal conclusion.

Write $s=e^{-\lambda t}$ for individual survival probability. Conditional on a clump of size $n$, its death probability is $(1-s{)}^n$. Averaging with the geometric distribution givesThusHere $0\le \theta <1$ and $t>0$; different clumps and experimental repetitions are understood independent for the sampling model.

The live-clump counts have independent binomial distributions, all with the same $p$. Up to a constant, their combined likelihood function is $p^R(1-p{)}^{N-R}$, maximized at $\hat{p}=R/N$ when $0<R<N$. Solving $p=s/(1-\theta +\theta s)$ for $s$ gives $s=p(1-\theta )/(1-\theta p)$. ThereforeMonotonicity of $p$ in $\lambda$ makes this the maximum-likelihood estimator. At the sampling boundaries, $R=N$ gives $\hat{\lambda }=0$, while $R=0$ gives only a supremum as $\lambda \to \infty$, with no finite maximizer. These are the limiting interpretations of the formula.

Taking the logarithm of the normalized probability density function, then summing over the independent observations, gives the log-likelihoodBoth $p$ and $\beta$ are parameters being inferred, so their contributions to the normalizing constant must be retained. The following R implementation validates the arguments and returns one number:

ll_f <- function(p, beta, x) {
  scalar <- function(v) {
    is.numeric(v) && !is.complex(v) && length(v) == 1L &&
      is.finite(v)
  }
  if (!scalar(p) || p <= 0 || p >= 1)
    stop("p must lie strictly between 0 and 1")
  if (!scalar(beta) || beta <= 0)
    stop("beta must be positive")
  if (!is.numeric(x) || is.complex(x) || !is.null(dim(x)) ||
      any(!is.finite(x)) || any(x <= 0))
    stop("x must be a vector of finite positive numbers")
  if (length(x) == 0L) return(0)
  t <- log1p(-p) - beta * x
  length(x) * (log(beta) - log(-log(p))) +
    sum(t - log(-expm1(t)))
}

Here $t_i=\log (1-p)-\beta x_i<0$ and -expm1(t) computes $1-e^t$ accurately near zero. The empty-data likelihood function is one, so its log-likelihood is zero. Invalid parameter values produce errors in this evaluator; the next algorithm rejects candidates outside the parameter support before calling it.

Define $R={\sum }_i(Y_i-\overline{Y}{)}^2$ and $s_Y^2=R/(n-1)$, with $n\ge 2$. The unrestricted maximum-likelihood estimates are $\hat{\mu }=\overline{Y}$ and ${\hat{\sigma }}^2=R/n$. Under the null hypothesis, maximizing the normal distribution's likelihood over the remaining variance givesbecause the cross term vanishes when ${\sum }_i(Y_i-\overline{Y})=0$. At either fitted variance, the exponential factor in the likelihood function is $e^{-n/2}$, so it cancels in the ratio:The standardized test statistic isThe sample variance $s_Y^2$ uses divisor $n-1$, whereas the maximum-likelihood estimate uses divisor $n$; this explains the $n-1$ in the final expression. Under a nondegenerate normal distribution, $R>0$ with probability one.

Let $L(\vartheta ;x)$ be the likelihood function, with full parameter space $\Theta$ and null hypothesis $\vartheta \in {\Theta }_0\subseteq \Theta$. The generalized likelihood-ratio test compares the best null fit to the best unrestricted fit:The likelihood ratio $R$ lies between zero and one, and small values count against the null hypothesis; equivalently, large values of the likelihood-ratio test statistic $\Lambda$ count against it.

Compute restricted and unrestricted maximum-likelihood estimates, evaluate their log-likelihoods, and form $\Lambda =2(\ell (\hat{\vartheta })-\ell ({\hat{\vartheta }}_0))$. Under the regular conditions of Wilks theorem, the null limit isThese statistical degrees of freedom count independent parameter restrictions. The regularity includes a locally identifiable smooth model, nonsingular Fisher information, and an interior null value; parameter-dependent support can invalidate the result.

For a chosen significance level $\alpha$, obtain the $(1-\alpha )$ quantile $q_{d,1-\alpha }$ of the chi-squared distribution and useEquivalently, compare the asymptotic p-value $\mathrm{Pr}({\chi }_d^2\ge {\Lambda }_{\mathrm{obs}})$ to $\alpha$. This is a large-sample calibration; part (c) requires direct treatment because its support depends on a parameter.

The count vector has a multinomial distribution, with likelihood functionMaximize its log-likelihood using a Lagrange multiplier for the sum constraint. The equations $n_i/{\pi }_i=\lambda$ and ${\sum }_i{\pi }_i=1$ give $\lambda =n$ and ${\hat{\pi }}_i=n_i/n$. Since all counts are positive and the log-likelihood is strictly concave on the interior of the probability simplex, this is the global maximum-likelihood estimate.

Under the uniform null hypothesis, each category has expected countThe common multinomial coefficient cancels from the likelihood ratio:Taking minus twice its logarithm gives exactlyThere are $k-1$ free category probabilities, because their sum is one, and the null fixes all of them. Thus its regular asymptotic chi-squared distribution has $k-1$ statistical degrees of freedom.