Fact-checked by Grok 2 weeks ago

Binomial distribution

The binomial distribution is a that describes the probability of achieving a specific number of successes in a fixed number of independent trials, where each trial has exactly two possible outcomes—typically labeled as "success" or "failure"—and the probability of success remains constant across all trials. It arises in scenarios modeled by a sequence of n trials, each with success probability p, and is fundamental to for counting events. The (PMF) of the distribution, denoted as B(n, p), gives the probability P(X = k) of exactly k successes in n trials as
P(X = k) = \binom{n}{k} p^k (1-p)^{n-k},
where \binom{n}{k} = \frac{n!}{k!(n-k)!} is the representing the number of ways to choose k successes out of n trials, for k = 0, 1, \dots, n. The parameters are n, a positive specifying the number of trials, and p, a between 0 and 1 indicating the success probability per trial. This formulation assumes independence between trials and a fixed success probability, distinguishing it from other distributions like the hypergeometric, which involves sampling without replacement.
Key statistical properties include the (mean) \mu = np, which represents the average number of successes, and the variance \sigma^2 = np(1-p), measuring the spread around the mean; the standard deviation is \sigma = \sqrt{np(1-p)}. The distribution is symmetric when p = 0.5, such as in flips, but skewed toward higher values of k when p > 0.5 or lower when p < 0.5. The mode, or most likely value of k, lies between \lfloor p(n+1) \rfloor and \lceil p(n+1) - 1 \rceil. In practice, the binomial distribution applies to fields like quality control (e.g., counting defective items in a batch), clinical trials (e.g., success rates of treatments), and surveys (e.g., proportion of favorable responses). For large n, it can be approximated by the when np and n(1-p) are both greater than 5 or 10, or by the when n is large and p is small such that \lambda = np is moderate. Historically, the binomial distribution was formalized by Jacob Bernoulli in his 1713 posthumous work Ars Conjectandi, building on earlier combinatorial ideas from Blaise Pascal in the 17th century related to the "problem of points." Abraham de Moivre later developed its normal approximation in 1733, advancing its role in statistical inference.

Definitions

Probability Mass Function

The binomial random variable X represents the number of successes in n independent Bernoulli trials, each with success probability p, where $0 \leq p \leq 1 and n is a positive integer. This distribution, originally developed by in his seminal 1713 work , models scenarios such as coin flips or quality control inspections with binary outcomes. The probability mass function (PMF) of X \sim \text{Bin}(n, p) is P(X = k) = \binom{n}{k} p^k (1-p)^{n-k}, where k = 0, 1, \dots, n and \binom{n}{k} = \frac{n!}{k!(n-k)!} denotes the binomial coefficient, counting the number of ways to select k successes from n trials. The PMF equals zero for any non-integer k outside this range. This formula derives from the independence of the trials: the probability of exactly k successes and n-k failures in a specific sequence is p^k (1-p)^{n-k}, and multiplying by the \binom{n}{k} possible sequences yields the total probability. Common notations include \text{Bin}(n, p) or B(n, p).

Parameters and Examples

The binomial distribution is characterized by two parameters: n, the number of independent , which must be a positive integer, and p, the probability of success on each trial, where $0 \leq p \leq 1. These parameters define the distribution's support over the integers k = 0, 1, \dots, n, representing the number of successes observed. This distribution models real-world scenarios involving a fixed number of independent trials, each with two possible outcomes: success or failure. For instance, it applies to coin flips, where a fair coin has p = 0.5; quality control processes, such as detecting defective items in manufacturing; or polling, where responses are yes/no on a binary question. Consider a fair coin tossed 10 times, so n = 10 and p = 0.5. The probability of exactly 5 heads, P(X = 5), is given by the probability mass function as \binom{10}{5} (0.5)^{10} = 252 / 1024 \approx 0.246, which is the mode and reflects the symmetric, bell-shaped distribution centered at the mean. In contrast, for a manufacturing process with a 20% defective rate (n = 100, p = 0.2), the probability mass function yields a right-skewed histogram, with higher probabilities clustered near fewer defectives (e.g., around 20) and a long tail toward more defectives, illustrating how small p shifts mass to the left. Boundary cases highlight the distribution's flexibility. When n = 1, it reduces to the with parameter p, focusing on a single trial's outcome. If p = 0, all probability mass concentrates at 0 (no successes); if p = 1, it concentrates at n (all successes), resulting in degenerate distributions.

Properties

Moments and Central Tendency

The binomial random variable X \sim \text{Bin}(n, p) represents the number of successes in n independent , each with success probability p. The expected value, or mean, of X is given by E[X] = np. This follows from expressing X as the sum of n indicator random variables Y_i for the i-th trial succeeding, where each Y_i \sim \text{Bernoulli}(p) has E[Y_i] = p, and applying the linearity of expectation: E[X] = E\left[\sum_{i=1}^n Y_i\right] = \sum_{i=1}^n E[Y_i] = np. The variance of X is \operatorname{Var}(X) = np(1-p), and the standard deviation is \sqrt{np(1-p)}. This derivation also uses the indicator representation, where \operatorname{Var}(Y_i) = p(1-p) for each trial, and independence implies \operatorname{Var}(X) = \sum_{i=1}^n \operatorname{Var}(Y_i) = np(1-p). Higher moments provide further characterization of the distribution's shape. The skewness, measuring asymmetry, is \frac{1-2p}{\sqrt{np(1-p)}}; it is zero when p = 0.5 (symmetric case), positive for p < 0.5, and negative for p > 0.5. The kurtosis, indicating tail heaviness relative to the normal distribution, is $3 + \frac{1-6p(1-p)}{np(1-p)}; the excess kurtosis term \frac{1-6p(1-p)}{np(1-p)} is positive for small np(1-p) and approaches zero as n grows large. The , the value of X with the highest probability, is \lfloor (n+1)p \rfloor or \lceil (n+1)p - 1 \rceil, with the being unimodal except near boundary values of p (e.g., 0 or 1) where it may be bimodal. The m, the value dividing the such that half the probability lies on each side, approximates np and lies within the [\lfloor np \rfloor, \lceil np \rceil] for outcomes. When p = 0.5, the is symmetric, so the , , and coincide at n/2. For p \neq 0.5, the skews: if p < 0.5, the exceeds the median and (right-skewed); if p > 0.5, the is less than the median and (left-skewed).

Tail Bounds and Inequalities

Tail bounds provide upper limits on the probability that a binomial random variable deviates significantly from its expected value, which is crucial for concentration inequalities and risk analysis. For a binomial random variable X \sim \operatorname{Bin}(n, p) with mean \mu = np, these bounds quantify the unlikelihood of extreme outcomes in the upper or lower tails. Markov's inequality offers a simple first-order bound for the upper tail. For a > np, it states that P(X \geq a) \leq \frac{np}{a}. This follows directly from applying the general Markov inequality to the non-negative random variable X, using its expectation. While loose, it provides a baseline without requiring higher moments or independence details beyond the mean. Hoeffding's inequality delivers a sharper, parameter-independent bound on deviations from the . Specifically, P(|X - np| \geq t) \leq 2 \exp\left(-\frac{2t^2}{n}\right) for t > 0. This result holds for the sum of bounded independent random variables, such as the Bernoulli trials underlying the binomial distribution, and relies on the range of each variable (here, 0 to 1). It excels in additive deviation scenarios, particularly when p is unknown or varies. Chernoff bounds provide tighter exponential control, especially for multiplicative deviations. For the upper tail, P(X \geq (1 + \delta)np) \leq \exp\left(-\frac{np \delta^2}{3}\right) holds for $0 < \delta < 1. A symmetric form applies to the lower tail: P(X \leq (1 - \delta)np) \leq \exp\left(-\frac{np \delta^2}{2}\right) for $0 < \delta < 1. These derive from optimizing the moment generating function via Markov's inequality on \mathbb{E}[e^{\lambda X}], yielding relative error bounds that scale with the variance np(1-p). Sanov's theorem extends to large deviations in the empirical distribution of binomial samples, stating that the probability of the sample proportion deviating from p by a fixed amount decays exponentially with rate given by the relative entropy D(\hat{p} \| p) = \hat{p} \log(\hat{p}/p) + (1 - \hat{p}) \log((1 - \hat{p})/(1 - p)). For binomial settings, this characterizes the precise asymptotics of tail events as n \to \infty, focusing on type deviations rather than fixed thresholds. These bounds find applications in algorithm analysis, such as bounding buffer overflow probabilities in randomized load balancing or controlling false positive rates in hypothesis testing for binomial outcomes. For instance, Chernoff bounds ensure that hashing algorithms achieve uniform distribution with high probability, limiting collision risks. Comparisons reveal trade-offs: Hoeffding's bound is simpler and p-independent, making it preferable for worst-case additive errors, but Chernoff's multiplicative form is tighter when deviations are proportional to the mean, especially for p near 0 or 1. Markov remains the coarsest, useful only for crude estimates.

Characterizing the Distribution

Cumulative Distribution Function

The cumulative distribution function (CDF) of a binomial random variable X with parameters n (number of trials) and p (success probability) is given by F(k; n, p) = P(X \leq k) = \sum_{i=0}^{\lfloor k \rfloor} \binom{n}{i} p^i (1-p)^{n-i}, where \binom{n}{i} denotes the binomial coefficient, and the sum is taken over integer values of i from 0 to the floor of k. This function is non-decreasing and right-continuous as a function of k, with boundary values F(-1; n, p) = 0 and F(n; n, p) = 1. It exhibits discontinuities (jumps) precisely at the integer points k = 0, 1, \dots, n, corresponding to the support of the distribution, where the size of each jump equals the probability mass at that point. An alternative expression for the CDF relates it to the regularized incomplete beta function I_x(a, b), defined as the ratio of the incomplete beta function to the complete beta function: F(k; n, p) = I_{1-p}(n - k, k + 1). This identity arises from integrating the beta density and equating it to the binomial sum via repeated integration by parts. For computational purposes, the direct summation is straightforward and accurate when n is small (e.g., n < 20), as it simply accumulates the terms of the probability mass function. However, for larger n, the summation can encounter numerical challenges, such as overflow from large intermediate binomial coefficients or loss of precision in floating-point arithmetic, particularly when p is close to 0 or 1. In modern software libraries, the beta function relation is often employed instead, leveraging stable algorithms for the incomplete beta (e.g., continued fraction expansions or series approximations) to ensure reliable evaluation across a wide range of parameters. In statistical applications, the binomial CDF plays a key role in hypothesis testing for proportions. For instance, under a null hypothesis specifying a particular p_0, the p-value for a one-sided lower-tail test based on an observed number of successes y is P(X \leq y \mid n, p_0) = F(y; n, p_0), providing the probability of observing a result at least as extreme as the data.

Quantile Function

The quantile function of the binomial distribution, denoted Q(α; n, p), is defined as the smallest integer k in {0, 1, ..., n} such that the cumulative distribution function F(k; n, p) ≥ α, where 0 < α < 1 and F(k; n, p) = ∑_{i=0}^k \binom{n}{i} p^i (1-p)^{n-i}. This function is non-decreasing in α and always returns an integer value within the support of the distribution. For α = 0.5, Q(0.5; n, p) gives the median, which lies near np. Computation of Q(α; n, p) typically involves a binary search over k from 0 to n, evaluating F(k; n, p) at midpoints until the smallest k satisfying the inequality is found; this approach is efficient given the monotonicity of the CDF and requires O(log n) evaluations. The CDF itself can be evaluated using its exact relation to the regularized incomplete beta function: F(k; n, p) = I_{1-p}(n - k, k + 1), where I_x(a, b) is the regularized incomplete beta function defined as I_x(a, b) = B_x(a, b) / B(a, b) with B_x(a, b) the incomplete beta function and B(a, b) the beta function. Equivalently, F(k; n, p) = 1 - I_p(k + 1, n - k), allowing the search condition to be rephrased as finding the minimal k where I_p(k + 1, n - k) ≤ 1 - α. Numerical implementations often use the inverse regularized incomplete beta function during the binary search to evaluate the CDF efficiently. Binary search remains the core method for precision across all parameter ranges. The quantile function finds application in determining critical values for binomial hypothesis tests, where Q(1 - α; n, p_0) under the null hypothesis p = p_0 defines the rejection region threshold. It also aids in sample size calculations to achieve specified power, by solving for n such that the quantile under the alternative hypothesis aligns with the desired type II error rate. For large n, an asymptotic approximation provides Q(α; n, p) ≈ np + z_α √[np(1 - p)], where z_α is the α-quantile of the standard normal distribution, though exact methods via the above procedures are preferred for small to moderate n.

Statistical Inference

Parameter Estimation

In the binomial distribution, parameter estimation typically arises in two scenarios: when the number of trials n is known and the success probability p must be estimated, or when both n and p are unknown. When n is fixed and known, the primary focus is on estimating p from an observed number of successes X in n independent Bernoulli trials. The maximum likelihood estimator (MLE) for p is given by \hat{p} = X / n, which represents the sample proportion of successes. This estimator is derived by maximizing the log-likelihood function \ell(p) = X \log p + (n - X) \log (1 - p), where the derivative with respect to p yields \frac{d\ell}{dp} = \frac{X}{p} - \frac{n - X}{1 - p} = 0, solving to \hat{p} = X / n. The MLE is unbiased, meaning E[\hat{p}] = p, and has variance \mathrm{Var}(\hat{p}) = p(1 - p)/n, which decreases as n increases. For large n, it achieves the minimum possible variance among unbiased estimators, as per the Cramér-Rao lower bound. The method of moments estimator for p coincides with the MLE when n is known, as it equates the sample mean \bar{X} = X/n to the population mean np, yielding the same \hat{p} = \bar{X}/n = X/n. When both n and p are unknown, estimation becomes more complex; the method of moments requires solving simultaneous equations from the first two population moments, \mu_1 = np and \mu_2 = np(1 - p) + (np)^2, using sample mean and variance, but this often leads to underestimation of n and numerical challenges due to the discrete nature of the parameters. In such cases, observed frequencies can provide initial estimates, though they are prone to bias, and alternatives like modeling with the —where n is interpreted as the number of failures until a fixed number of successes—may be more suitable for certain data structures.

Confidence Intervals for p

Confidence intervals for the binomial parameter p, with n known, provide a range [L, U] such that the probability P(L \leq p \leq U \mid \text{data}) \approx 1 - \alpha. These intervals quantify uncertainty around the point estimate \hat{p} = X/n, where X is the observed number of successes. Various methods exist to construct such intervals, balancing coverage probability (the actual probability that the interval contains the true p) and interval width, with performance varying by sample size and true p value. The Wald interval, also known as the normal approximation interval, is given by \hat{p} \pm z_{\alpha/2} \sqrt{\frac{\hat{p}(1 - \hat{p})}{n}}, where z_{\alpha/2} is the (1 - \alpha/2)-quantile of the standard normal distribution. This method relies on the asymptotic normality of \hat{p} and performs well for large n and p away from 0 or 1, but it exhibits poor coverage for small n or when p is near the boundaries, often undercovering the true p by up to 10-20% in simulations. The Wilson score interval improves upon the Wald by centering on an adjusted estimate and is formulated as \hat{p} + \frac{z_{\alpha/2}^2}{2n} \pm z_{\alpha/2} \sqrt{ \frac{\hat{p}(1 - \hat{p})/n + z_{\alpha/2}^2/(4n^2)}{1 + z_{\alpha/2}^2/n} }, where the center shifts toward 0.5 for better symmetry. Derived from inverting the score test, it achieves near-nominal coverage across a wider range of n and p, with simulation studies showing coverage probabilities closer to $1 - \alpha than the Wald, especially for n < 50. This method was originally proposed for practical inference in binomial settings. The Agresti-Coull interval simplifies computation by adding pseudocounts: treat the data as if there were X + z_{\alpha/2}^2/2 successes in n + z_{\alpha/2}^2 trials (often z_{\alpha/2}^2 = 4 for 95% confidence), then apply the to these adjusted values. For 95% intervals, this equates to adding 2 successes and 4 trials. It offers good coverage for small samples, outperforming the exact in terms of shorter average length while maintaining coverage above 94%, and is recommended for its ease and reliability in applied settings. The arcsine transformation interval applies the variance-stabilizing transformation \sin^2(\arcsin(\sqrt{\hat{p}}) \pm z_{\alpha/2} / (2 \sqrt{n})), which stretches the distribution ends to improve normality approximation. This method stabilizes the variance of \hat{p} and yields reasonable coverage for moderate n, though it can produce asymmetric intervals and performs less favorably near boundaries compared to or in simulation-based evaluations. Among these, the Wilson score interval is often preferred for its balance of coverage accuracy and computational simplicity, with coverage rates typically within 1-2% of nominal across diverse scenarios, as evidenced by extensive simulations. The exact Clopper-Pearson interval, obtained by inverting the binomial test using beta quantiles, guarantees at least $1 - \alpha coverage but tends to be overly conservative (coverage up to 5-10% above nominal) and computationally intensive for large n, making it suitable mainly for small samples. Comparative studies recommend Wilson or Agresti-Coull over Wald for most practical uses, with coverage assessed via Monte Carlo simulations showing Wilson superior for p near 0 or 1.

Exact Relations

The binomial distribution with a single trial, denoted \text{Bin}(1, p), is identical to the with success probability p, where the probability mass function (PMF) is \Pr(X = 1) = p and \Pr(X = 0) = 1 - p. More generally, a binomial random variable X \sim \text{Bin}(n, p) arises as the sum of n independent and identically distributed (i.i.d.) , each with success probability p; the PMF of the binomial is then \Pr(X = k) = \binom{n}{k} p^k (1-p)^{n-k} for k = 0, 1, \dots, n. The binomial distribution is closed under summation for independent random variables sharing the same success probability. Specifically, if X_i \sim \text{Bin}(n_i, p) for i = 1, \dots, m are independent, then their sum S = \sum_{i=1}^m X_i follows \text{Bin}\left( \sum_{i=1}^m n_i, p \right), with PMF \Pr(S = k) = \binom{\sum n_i}{k} p^k (1-p)^{\sum n_i - k} for k = 0, 1, \dots, \sum n_i. This property follows directly from the convolution of the individual PMFs, leveraging the identical p to preserve the binomial form. A generalization occurs when the success probabilities differ across trials: the sum of independent Bernoulli random variables X_i \sim \text{Bern}(p_i) for i = 1, \dots, n follows the , with PMF given by the convolution \Pr(S = k) = \sum_{A \subseteq : |A|=k} \prod_{i \in A} p_i \prod_{j \notin A} (1 - p_j). When all p_i = p, the Poisson binomial reduces exactly to the \text{Bin}(n, p). This relation highlights the binomial as a special homogeneous case of the more general Poisson binomial. The beta-binomial distribution emerges when the binomial success probability is uncertain and follows a beta prior. If p \sim \text{Beta}(\alpha, \beta) and, conditional on p, X \mid p \sim \text{Bin}(n, p), then the marginal distribution of X is beta-binomial with parameters n, \alpha, \beta, having PMF \Pr(X = k) = \binom{n}{k} \frac{B(k + \alpha, n - k + \beta)}{B(\alpha, \beta)}, where B(\cdot, \cdot) is the beta function. This compound distribution accounts for heterogeneity in p, leading to overdispersion relative to the standard (variance np(1-p) \frac{\alpha + \beta + n}{\alpha + \beta + 1} > np(1-p)). The binomial distribution relates to the through a limiting process that preserves exact structure in the infinite-population case. The \text{Hyper}(N, K, n) models the number of successes in n draws without replacement from a finite of N with K successes, with PMF \Pr(X = k) = \frac{\binom{K}{k} \binom{N-K}{n-k}}{\binom{N}{n}}. As N \to \infty with K/N \to p fixed, the hypergeometric converges exactly to the binomial \text{Bin}(n, p), reflecting independent trials in an effectively infinite . The negative binomial distribution inverts the fixed-trials perspective of the binomial. While the binomial \text{Bin}(n, p) counts successes in n fixed trials, the negative binomial \text{NB}(r, p) (number of failures before the r-th success) counts trials until r fixed successes, with PMF \Pr(X = k) = \binom{k + r - 1}{k} p^r (1-p)^k for k = 0, 1, \dots. The two are inversely related: for fixed n and r, the event \{ \text{NB}(r, p) \geq n \} corresponds to at most r-1 successes in the first n + r - 1 trials under the binomial distribution Bin(n + r - 1, p), i.e., linking their and via P(\text{NB}(r, p) \geq n) = P(\text{Bin}(n + r - 1, p) \leq r - 1).

Limiting Approximations

The binomial distribution arises as the sum of independent Bernoulli trials, and under certain limiting regimes, it converges in distribution to other well-known distributions, providing useful approximations for computation and analysis. One primary limiting case occurs when the number of trials n tends to infinity while the success probability p tends to zero such that the product \lambda = np remains fixed and positive. In this regime, known as the law of rare events, the binomial distribution \mathrm{Bin}(n, p) converges to the Poisson distribution with parameter \lambda. This Poisson approximation is particularly effective for modeling , where successes are infrequent relative to the total trials. The (PMF) of the , given by P(X = k) = \binom{n}{k} p^k (1-p)^{n-k}, approximates the Poisson PMF P(Y = k) = e^{-\lambda} \lambda^k / k! through the expansion \binom{n}{k} \left(\frac{\lambda}{n}\right)^k \left(1 - \frac{\lambda}{n}\right)^{n-k} \approx \frac{e^{-\lambda} \lambda^k}{k!}, as n \to \infty, with the approximation improving when n is large and p is small (typically n \geq 100 and np \leq 10). The error in this approximation can be bounded using techniques like the Stein-Chen method, yielding distances on the order of \min(1, 1/\lambda) p, which decreases as p \to 0. This limit is foundational for applications in rare event simulation and , where exact binomial computations become infeasible. A second key limiting approximation arises when n \to \infty with p fixed (or more generally, $0 < p < 1), leading to convergence to the via the (CLT). Specifically, if X \sim \mathrm{Bin}(n, p), then the standardized variable (X - np)/\sqrt{np(1-p)} converges in distribution to the standard normal Z \sim N(0,1), so X \approx N(np, np(1-p)) for large n. This , originally established for the binomial case, requires the conditions np \to \infty and n(1-p) \to \infty to ensure the approximation's validity, typically holding well when n \geq 30 and p is not too close to 0 or 1. The is advantageous for moderate p and large n, facilitating analytical tractability in and . For cumulative probabilities, the continuity correction enhances accuracy by treating the discrete binomial as continuous: P(a \leq X \leq b) \approx \Phi\left(\frac{b + 0.5 - np}{\sqrt{np(1-p)}}\right) - \Phi\left(\frac{a - 0.5 - np}{\sqrt{np(1-p)}}\right), where \Phi is the standard normal CDF. The error in the normal approximation is quantified by the Berry–Esseen theorem, which provides a uniform bound on the Kolmogorov distance of O(1/\sqrt{n}), specifically C \rho / (\sigma^3 \sqrt{n}) with C \approx 0.4748 for the optimal constant, where \rho = E[|X_1 - p|^3] and \sigma^2 = p(1-p) for the underlying Bernoulli variables. This rate confirms the approximation's asymptotic sharpness, though practical errors may be smaller for symmetric cases near p=0.5. The local central limit theorem extends this to point probabilities, stating that for lattice distributions like the binomial, the local densities converge uniformly to the normal density: P(X = k) \approx \frac{1}{\sqrt{2\pi np(1-p)}} \exp\left( -\frac{(k - np)^2}{2np(1-p)} \right) as n \to \infty, under the same conditions, with error terms also O(1/\sqrt{n}). Overall, the Poisson limit suits sparse success scenarios (e.g., defect counts in manufacturing), while the normal excels in balanced, high-volume settings (e.g., election polling), guiding the choice based on np and n(1-p).

Computational Aspects

Generating Random Variates

Generating random variates from the binomial distribution Bin(n, p) is essential for Monte Carlo simulations and statistical modeling. Common methods include exact techniques like the inverse transform sampling and the summation of Bernoulli trials, as well as rejection-based approaches for efficiency. Special-purpose algorithms optimize performance for specific parameter ranges, while software libraries provide robust implementations. The inverse transform method generates a binomial variate by drawing a uniform random variable U ~ Uniform(0,1) and finding the smallest integer k such that the cumulative distribution function F(k) ≥ U, where F(k) = ∑_{i=0}^k \binom{n}{i} p^i (1-p)^{n-i}. This can be computed efficiently for small n using a recursive approach that accumulates probabilities sequentially until the sum exceeds U, avoiding full computation of the binomial coefficients each time. The expected time complexity is O(np) in the worst case but often better due to early termination on average. A simple exact method, equivalent to inverse transform in some implementations, involves generating n independent Bernoulli(p) random variables (each via a single uniform comparison to p) and summing them to obtain the binomial variate; this requires O(n) time and is straightforward for moderate n. Rejection sampling enhances efficiency, particularly when approximations are viable. For cases where np is small (e.g., p near 0 and n moderate), a Poisson(λ = np) proposal distribution can be used, generating a Poisson variate and accepting it if it is at most n; rejections are rare under the approximation. More generally, rejection methods employ envelopes like triangular or exponential functions to bound the probability mass function, accepting or rejecting based on a uniform draw. The Binomial Test Point Envelope (BTPE) algorithm is a specialized rejection sampler for moderate n where min(np, n(1-p)) ≥ 10. It decomposes the support into regions (triangular central, parallelogram sides, exponential tails) with a piecewise majorizing function, using 2-4 uniform variates per accepted sample on average and fixed memory independent of n. BTPE outperforms earlier constant-memory methods by a factor of over 2 in execution time for its parameter range. The alias method provides fast generation for discrete distributions with finite support like the binomial. It preprocesses the probabilities into probability and alias tables in O(n) time, allowing O(1) time per variate thereafter by drawing a uniform index and a sub-uniform to decide between the index or its alias. This is particularly efficient for repeated sampling with fixed n and p. For large n, approximate methods improve efficiency; for example, rejection sampling from a normal approximation N(np, np(1-p))—generating a normal variate, rounding to the nearest integer k (0 ≤ k ≤ n), and accepting with probability proportional to the ratio of the exact PMF to the normal density—yields exact variates with expected O(1) uniforms per sample after occasional rejections. The inverse CDF can reference the cumulative distribution function for quantile-based generation in small-n cases. In software, R's rbinom(n, size, prob) function generates n binomial variates with size trials and success probability prob, using a combination of methods including BTPE for suitable parameters to ensure efficiency. Similarly, NumPy's numpy.random.Generator.binomial(n, p, size=None) (recommended over the legacy numpy.random.binomial) draws samples (scalar or array of shape size) from Bin(n, p), suitable for use when np ≤ 5 for estimating the standard error of a proportion, and internally employs optimized algorithms like those in the PCG family for uniforms. Direct methods like Bernoulli summation run in O(n) time, while rejection-based algorithms like BTPE achieve near-constant time per variate (average 2-4 uniforms), making them preferable for large-scale simulations. For validation, simulated samples should have empirical mean ≈ np and variance ≈ np(1-p), matching theoretical moments; deviations indicate implementation issues.

Numerical Evaluation

The probability mass function (PMF) of the binomial distribution can be computed via direct summation using the formula P(X = k) = \binom{n}{k} p^k (1-p)^{n-k}, where the binomial coefficient \binom{n}{k} is calculated recursively to minimize intermediate overflow: \binom{n}{k} = \binom{n}{k-1} \cdot \frac{n-k+1}{k}, starting from \binom{n}{0} = 1. This multiplicative approach avoids computing large factorials directly and is suitable for moderate n (up to around 1000 in double-precision floating-point arithmetic). For larger n, direct computation risks overflow and underflow in floating-point representations, leading to numerical instability; relative errors can exceed machine epsilon (about $10^{-16} for double precision) when intermediate terms grow beyond $2^{53}, the mantissa limit. To address this, the logarithm of the PMF is often evaluated instead: \log P(X = k) = \log \binom{n}{k} + k \log p + (n-k) \log(1-p), where \log \binom{n}{k} uses the log-gamma function via \log \Gamma(n+1) - \log \Gamma(k+1) - \log \Gamma(n-k+1), implemented accurately in numerical libraries without explicit for the core computation. , \log n! \approx n \log n - n + \frac{1}{2} \log (2 \pi n), provides a fallback for extremely large n beyond library support, with relative errors decreasing as O(1/n). The cumulative distribution function (CDF) F(k; n, p) = \sum_{i=0}^{k} P(X = i) can be computed by summing the recursive or log PMF terms, but for efficiency and precision with large n, it leverages the relation to the regularized incomplete beta function: F(k; n, p) = 1 - I_p(k+1, n-k), where I_x(a,b) is evaluated using continued fraction expansions or numerical quadrature to avoid summation loops and overflow. These methods ensure relative errors below $10^{-15} for n > 1000 in standard double precision, switching to approximations only when exact computation exceeds feasible precision. Quantiles, or the inverse CDF, are computed numerically by solving F(q; n, p) = \alpha for q, typically using the Newton-Raphson method on the CDF derivative (the PMF) for rapid convergence or search for guaranteed monotonicity and robustness, converging in O(\log n) iterations with errors bounded by 1 in the domain. Standard libraries implement these techniques for practical evaluation. In Python's , binom.pmf(k, n, p) and binom.cdf(k, n, p) use log-gamma for PMF and incomplete for CDF, while binom.ppf(alpha, n, p) employs ; for arbitrary precision with large n, the mpmath library supports exact computation via multi-precision gamma functions. Similarly, MATLAB's binopdf(k, n, p) and binocdf(k, n, p) rely on log-gamma and integrals, handling n up to machine limits with relative errors under $10^{-14}. Error analysis in these implementations confirms that floating-point relative errors remain below $10^{-13} for n \leq 10^6 and p away from 0 or 1, prompting a shift to or approximations beyond that threshold for stability.

Applications

Common Uses in Probability and Statistics

The binomial distribution is frequently employed in hypothesis testing to assess whether an observed proportion of successes in a fixed number of independent trials deviates significantly from a hypothesized p_0. For instance, in testing the fairness of a , the H_0: p = 0.5 can be evaluated using the exact , where the is calculated as the sum of the probabilities in the tail(s) of the distribution beyond the observed number of heads. In quality control, the binomial model underpins procedures, where a random sample of size n is inspected from a lot, and the lot is rejected if the number of defectives X exceeds a c, thereby controlling defect rates based on the assumed probability p of a defective item. This approach balances producer's and consumer's risks by designing operating characteristic (OC) curves that specify acceptance probabilities for given quality levels, such as (AQL) and lot tolerance percent defective (LTPD). Polling and surveys commonly use the binomial distribution to estimate voter support or other outcomes, where the sample proportion \hat{p} = X/n from affirmative responses in a sample of size n serves as an for the p. This facilitates about , such as the favoring a , with the underlying structure ensuring the estimator's properties under independent responses. In risk analysis, the binomial distribution quantifies the reliability of redundant systems by computing the probability of k or more failures among n independent components, each with failure probability p, which informs design decisions in safety-critical applications like . For example, tables of required component reliabilities for specified system reliabilities are derived from binomial probabilities to optimize levels. Educational examples in often invoke the binomial distribution to model , such as the probability of observing a dominant in offspring from heterozygous parents, where p = 0.75 reflects the genotypic ratios (3:1 dominant to recessive). This illustrates probabilistic expectations in dihybrid crosses under assortment. Despite its versatility, the binomial distribution assumes trial and a fixed success probability p; violations, such as dependence between trials or varying p, necessitate alternative models to avoid biased inferences. and intervals for p, as discussed in , build on these foundations but require verification of the assumptions.

Extensions in Modeling

The extends the standard model by incorporating , where the success probability p is treated as a drawn from a with shape parameters \alpha > 0 and \beta > 0. This hierarchical approach accounts for extra variation in the data beyond what independent identically distributed (i.i.d.) trials would produce, such as when trials are correlated due to unobserved heterogeneity. The mean remains E(X) = n \frac{\alpha}{\alpha + \beta}, but the variance is inflated to \operatorname{Var}(X) = n \frac{\alpha \beta}{(\alpha + \beta)^2} \cdot \frac{\alpha + \beta + n}{\alpha + \beta + 1}, which can be rewritten as np(1-p)(1 + (n-1)\rho) where p = \frac{\alpha}{\alpha + \beta} and \rho = \frac{1}{\alpha + \beta + 1} measures the degree of (with \rho > 0 indicating positive correlation among trials). Binomial logistic regression further generalizes the binomial distribution by allowing the success probability p_i to vary across observations based on predictors, modeled via the link function: p_i = \frac{1}{1 + \exp(-X_i \beta)}, where X_i is a of covariates and \beta are the coefficients. This framework accommodates non-constant p in settings while assuming i.i.d. trials conditional on the predictors, enabling on how factors like environmental variables influence outcomes. The zero-inflated distribution addresses scenarios with excess zeros not adequately captured by the standard , such as structural zeros from a separate process alongside variability. It is defined as a : with probability \omega (0 ≤ \omega < 1), the outcome is zero (degenerate at 0), and with probability $1 - \omega, it follows a (n, \pi) distribution. This model is particularly useful for , like counts in biological assays where many trials yield no successes due to inherent impossibilities. For dependent trials, correlated binomial models employ copulas or multivariate extensions to introduce joint dependencies while preserving marginal binomial structures. Pair-copula constructions, such as D-vine models with Gaussian or Clayton copulas, build higher-dimensional distributions by linking bivariate copulas, allowing specification of marginal probabilities and pairwise correlations \rho. These approaches flexibly capture clustering or serial dependence in binary data without assuming i.i.d. trials. Such extensions find applications in , where beta-binomial models handle in count data or rates, such as estimating hatching probabilities in animal clutches affected by environmental heterogeneity. In , they model clustered default rates, using beta-binomial or correlated binomial frameworks to account for unobserved factors driving simultaneous defaults in loan portfolios, improving over independent assumptions. In contrast to the beta-binomial, which assumes a common but uncertain p leading to , the applies to independent trials with heterogeneous success probabilities p_i, modeling the sum without introducing correlation.

History and Development

The roots of the binomial distribution trace back to early combinatorial problems in ancient texts, such as the Sūtra from ancient (around 600 BCE) and works by Greek mathematicians like Pappus of Alexandria in the , which explored counting combinations and recursive patterns. In the 14th century, Jewish scholar Levi ben Gerson advanced these ideas by examining proportions in sample spaces. During the 16th century, the binomial coefficients gained prominence through the study of , with contributions from mathematicians like Michael Stifel. The "problem of points," concerning the fair division of stakes in interrupted games, was addressed by Italian scholars (1494) and (1539), who introduced concepts of probability and . In 1654, and corresponded on this problem, developing methods for equal probabilities (p = 0.5) using combinatorial enumeration. further generalized in his 1657 De Ratiociniis in Ludo Aleae. The formalization of the binomial distribution occurred in the early . Jacob Bernoulli, in his posthumously published Ars Conjectandi (1713), derived the for any success probability p, linking it to the and providing the first rigorous treatment beyond fair odds. Abraham de Moivre extended this work in 1733 (published in Approximatio ad summam terminorum binomii (a + b)^n in serie expansi), developing the normal approximation to the binomial distribution, which facilitated calculations for large n. In the 19th and 20th centuries, the distribution's applications expanded in statistics and . and others refined its use in . Notably, in 1936, applied the binomial model to Gregor Mendel's 1866 pea plant experiments, revealing data that fit the expected ratios too closely (e.g., for n=8,023 trials with p=0.75, observed 6,022 yellow peas versus expected 6,017), suggesting possible data adjustment and sparking debates on scientific integrity.

References

  1. [1]
    1.3.6.6.18. Binomial Distribution - Information Technology Laboratory
    Probability Mass Function, The binomial distribution is used when there are exactly two mutually exclusive outcomes of a trial.
  2. [2]
    Binomial distribution - Student Academic Success - Monash University
    The binomial. distribution is a key concept in probability. that models situations where you repeat the same experiment several times, and each time there are ...Bernoulli sequence · The binomial distribution · Expectation of the binomial...
  3. [3]
    Binomial Distribution
    In statistics the so-called binomial distribution describes the possible number of times that a particular event will occur in a sequence of observations.
  4. [4]
    Random Variables
    Blaise Pascal (1623-1662) used binomial coefficients and an informal use of the binomial distribution to help him solve the problem of points. Thus, some ...
  5. [5]
    The Binomial Distribution
    The binomial distribution for a random variable X with parameters n and p represents the sum of n independent variables Z which may assume the values 0 or 1.
  6. [6]
    The significance of Jacob Bernoulli's Ars Conjectandi for the ...
    The theme of this article is that Jacob's contributions to the philosophy of probability are still fresh in many ways.
  7. [7]
    Proof: Probability mass function of the binomial distribution
    May 11, 2020 · Proof: A binomial variable is defined as the number of successes observed in n n independent trials, where each trial has two possible ...
  8. [8]
  9. [9]
    [PDF] Chapter 3: Discrete Random Variables And Probability
    independent Bernoulli trials, such as the number of heads in ten coin flips. The binomial distribution is frequently used in quality control and opinion polls.
  10. [10]
    10.2 - Is X Binomial? | STAT 414 - STAT ONLINE
    A discrete random variable is a binomial random variable if an experiment, or trial, is performed in exactly the same way times.Missing: flips | Show results with:flips
  11. [11]
    10.4 - Effect of n and p on Shape | STAT 414
    The shape of the binomial distribution is directly related, and not surprisingly, to two things: For small p and small n, the binomial distribution is what we ...
  12. [12]
    [PDF] Stat 8931 (Aster Models) Lecture Slides Deck 9 [1ex] Directions of ...
    Degeneracy. Binomial(n,p) distributions with p = 0 or p = 1 are degenerate. p = 0 implies X = 0 with probability one. p = 1 implies X = n with probability one.
  13. [13]
    Markov and Chebyshev Inequalities - Probability Course
    Note that X is a nonnegative random variable and EX=np. Applying Markov's inequality, we obtain P(X≥αn)≤EXαn=pnαn=pα. For p=12 and α=34, we obtain P(X≥3n4)≤23.
  14. [14]
    [PDF] probability inequalities for sums of bounded - UMBC CSEE
    This is known as Jensen's inequality. In this section probability bounds for sums of independent random variables are stated and discussed. The proofs are ...
  15. [15]
    [PDF] Chernoff-Hoeffding Bounds for Applications with - Cornell eCommons
    Chernoff-Hoeffding bounds are fundamental tools used in bounding the tail probabilities of the sums of bounded and independent random variables.<|separator|>
  16. [16]
    [PDF] A survey of Chernoff and Hoeffding bounds - arXiv
    Jun 19, 2025 · It is similar to the one in [2] for the binomial case. The following inequality will be used. Lemma 6 For every 0 ≤ δ ≤ 1 we have the following.
  17. [17]
    [PDF] ECE 302: Lecture 4.3 Cumulative Distribution Function
    CDF must satisfy these properties: Non-decreasing, FX (−∞) = 0, and FX (∞) = 1. P[a ≤ X ≤ b] = FX (b) − FX (a). Right continuous: Solid dot on at the start.Missing: binomial | Show results with:binomial
  18. [18]
    (PDF) Derivatives of the Incomplete Beta Function - ResearchGate
    Aug 10, 2025 · In Section 3 the computation of the binomial cdf uses the relation with the regularized incomplete beta function where B(x 1 , n 1 , p) = I ...
  19. [19]
    [PDF] Some Tests Based on the Binomial Distribution (Conover Chapter ...
    For the P-value of an upper one-sided hypothesis test, we just take. P = P(Y ≥ y|p = p∗)=1 − B(y − 1;n, p∗). (1.7). Again, only certain specific P-values are ...
  20. [20]
    [PDF] Computing the Cumulative Distribution Function and Quantiles of ...
    Feb 20, 2018 · The binomial coefficient computation contributes (2j − 1)δ× ... , Formula 33 is definitely usable, for computing the CDF/SF and inverse CDF/SF.
  21. [21]
    1.4 - Method of Moments | STAT 415 - STAT ONLINE
    The method of moments involves equating sample moments with theoretical moments. So, let's start by making sure we recall the definitions of theoretical ...
  22. [22]
    [PDF] Estimation of Binomial Parameters when Both n, p are Unknown
    Feb 8, 2004 · The underestimation is generally so drastic as to make many of the common estimates essen- tially useless.
  23. [23]
    1.5 - Maximum Likelihood Estimation | STAT 504
    The likelihood function is essentially the distribution of a random variable (or joint distribution of all values if a sample of the random variable is ...
  24. [24]
    [PDF] Maximum Likelihood Estimation
    Step one of MLE is to write the likelihood of a Bernoulli as a function that we can maximize. Since a Bernoulli is a discrete distribution, the likelihood is ...
  25. [25]
    [PDF] WILD 502: Binomial Likelihood – page 1
    WILD 502: Binomial Likelihood – page 3. Maximum Likelihood Estimation – the Binomial Distribution. This is all very good if you are working in a situation ...
  26. [26]
    [PDF] Lecture 6: Estimators 6.1 Method of moments estimator
    → 0. Example 1: Binomial Distribution. Assume that we obtain a single observation Y ∼ Bin(n, p), and we assume that n is known. The goal is to estimate p.
  27. [27]
    [PDF] Method-of-Moments Estimation - MIT OpenCourseWare
    In the next series lectures we will study three methods of estimation: method-of- moments, maximum likelihood and Bayesian estimation.
  28. [28]
    Two-Stage and Sequential Unbiased Estimation of N in Binomial ...
    Dec 31, 2021 · We propose two-stage and sequential procedures to estimate the unknown parameter N of a binomial distribution with unknown parameter p, when we ...
  29. [29]
    Interval Estimation for a Binomial Proportion - Project Euclid
    May 2001 Interval Estimation for a Binomial Proportion. Lawrence D. Brown, T. Tony Cai, Anirban DasGupta · DOWNLOAD PDF + SAVE TO MY LIBRARY. Statist. Sci. 16(2): ...
  30. [30]
    Approximate is Better than “Exact” for Interval Estimation of Binomial ...
    Mar 22, 2012 · For interval estimation of a proportion, coverage probabilities tend to be too large for “exact” confidence intervals based on inverting the binomial test.
  31. [31]
    Special Distributions | Bernoulli Distribution | Binomial Distribution
    The indicator random variable for an event A has Bernoulli distribution with parameter p=P(A), so we can write IA∼Bernoulli(P(A)). Geometric Distribution. The ...
  32. [32]
    4 Standard discrete distributions | Distribution Theory - Bookdown
    Oct 28, 2024 · A binomial distribution is used to model the number of successes in n n independent Bernoulli trials (Def. 4.5). A simple example is tossing a ...
  33. [33]
    [PDF] Learning Poisson Binomial Distributions - CS@Columbia
    Dec 7, 2013 · A Poisson Binomial Distribution (PBD) over {0, 1,...,n} is the distribution of a sum of n independent. Bernoulli random variables which may have ...
  34. [34]
    [PDF] Chapter 1 Special Distributions
    Exercise 1.3 If X ∼ Binomial(m, p) and Y ∼ Binomial(n, p) are independent, then the conditional distribution of X given that X + Y = N is Hypergeometric(m + n,N ...
  35. [35]
    [PDF] The Beta-Binomial Distribution - Cornell eCommons
    The beta function can be written as B(m, n) = (m−1)!(n−1)!. (m+n−1)!. , which allows us to rewrite the previous equation as. Pr (n, k) = n k. B(k + α, n − k + β).
  36. [36]
    [PDF] The Hypergeometric Distribution
    2 The Binomial Distribution as a Limit of Hypergeometric Distributions. The connection between hypergeometric and binomial distributions is to the level of the.
  37. [37]
    The Poisson Approximation to the Poisson Binomial Distribution - jstor
    The limit theorem of von Mises suggests that the Poisson approximation will be reliable provided that n is large, a is small, and X is moderate. But even.Missing: citation | Show results with:citation
  38. [38]
    On the Convergence of Binomial to Poisson Distributions
    The purpose of this note is to demonstrate, with an elementary argument, that the binomial distributions converge very strongly to the Poisson distributions.
  39. [39]
    Deriving the central limit theorem from the de Moivre–Laplace theorem
    The de Moivre–Laplace theorem is a special case of the central limit theorem for Bernoulli random variables, and can be proved by direct computation.
  40. [40]
    28.1 - Normal Approximation to Binomial | STAT 414
    We will now focus on using the normal distribution to approximate binomial probabilities. The Central Limit Theorem is the tool that allows us to do so.
  41. [41]
    [PDF] On a bound of the absolute constant in the Berry–Esseen inequality ...
    Sep 14, 2018 · Abstract It is shown that the absolute constant in the Berry–Esseen inequality for i.i.d. Bernoulli random variables is strictly less than the ...
  42. [42]
    Berry-Esseen bound for binomial distribution - Math Stack Exchange
    Jun 20, 2015 · Theorem 2 on page 3 gathers together the results in the paper to show that C can be taken to be .4215 in general.Estimates for the normal approximation of the binomial distributionNormal approximation of tail probability in binomial distributionMore results from math.stackexchange.com
  43. [43]
    [PDF] An Elementary Proof of the Local Central Limit Theorem
    To keep the notation simple, we first state and prove a local limit theorem for sequences of random variables. At the end of this section we state a more ...
  44. [44]
  45. [45]
    [PDF] 1 Inverse Transform Method
    Suppose we want X to have a binomial (n, p) distribution, that is, p(k) = P(X = k) = n k pk(1 − p)n−k, 0 ≤ k ≤ n. One could, in principle, use the ...
  46. [46]
    [PDF] 7.3 Rejection Method: Gamma, Poisson, Binomial Deviates
    The rejection method is a powerful, general technique for generating random deviates whose distribution functionp(x)dx(probability of a value occurring between ...
  47. [47]
    Binomial random variate generation - ACM Digital Library
    The most basic Bernoulli method generates n independent U(0, 1) random variates, and returns the numbers that are less than or equal to p. Algorithm BU. 1.
  48. [48]
    An Analysis of an Open Source Binomial Random Variate ... - MDPI
    Oct 26, 2023 · BTPE uses rejection sampling, and BTPE's authors originally provided an analytical formula for the expected number of iterations in terms of n ...
  49. [49]
    Binomial function - RDocumentation
    The length of the result is determined by n for rbinom , and is the maximum of the lengths of the numerical arguments for the other functions. The numerical ...Missing: implementation | Show results with:implementation
  50. [50]
    numpy.random.binomial — NumPy v2.4.dev0 Manual
    Draw samples from a binomial distribution. Samples are drawn from a binomial distribution with specified parameters, n trials and p probability of success where ...
  51. [51]
    [PDF] Fast and Accurate Computation of Binomial Probabilities - R Project
    Feb 5, 2002 · Numerical results and timings are presented in Section 4. Since n! = Γ(n + 1) and the log-gamma function is provided in most math libraries, ...Missing: pmf | Show results with:pmf
  52. [52]
    [PDF] stirling's formula - keith conrad
    Stirling's formula can also be expressed as an estimate for log(n!): (1.1) log(n!) = nlog n − n + 1 2 log n + 1 2 log(2π) + εn, where εn → 0 as n → ∞.
  53. [53]
    scipy.stats.binom — SciPy v1.16.2 Manual
    A binomial discrete random variable. As an instance of the rv_discrete class, binom object inherits from it a collection of generic methods.1.15.2 · 1.15.1 · 1.12.0 · 1.13.1
  54. [54]
    8.1.2 - Hypothesis Testing | STAT 200
    If n p 0 < 10 or n ( 1 − p 0 ) < 10 then the distribution of sample proportions follows a binomial distribution. We will not be conducting this test by hand in ...Missing: cdf | Show results with:cdf
  55. [55]
    Binomial Test - University of Texas at Austin
    A binomial test uses sample data to determine if the population proportion of one level in a binary (or dichotomous) variable equals a specific claimed value.
  56. [56]
    6.2.2.2. Choosing a Sampling Plan with a given OC Curve
    To design a sampling plan with a given OC curve, two points are needed, such as AQL and LTPD, or Producer's and Consumer's Risk. Sample size and acceptance ...
  57. [57]
    [PDF] Statistical Methods for Quality Control I. Acceptance Sampling
    Acceptance sampling decides whether to accept or reject shipments by selecting, testing, and deciding based on the proportion of defective items.
  58. [58]
    Chapter 13 The Binomial Distribution | Introduction to Statistics and ...
    If you flip a coin repeatedly, say 10 times, and count up the number of heads, this number is drawn from what's called a binomial distribution.
  59. [59]
    [PDF] Math C067 Sampling Distributions - Temple CIS
    Examples of Sampling. • A pollster may try to estimate the proportion of voters in favor of a particular presiden- tial candidate by polling a random sample ...
  60. [60]
    [PDF] Tables of component reliability for binomial redundancy applications
    Extensive tables of component reliabilities required for a specified system reliability are presented for systems whose reliability is expressible by the ...
  61. [61]
    [PDF] RELIABILITY MODEL AND ASSESSMENT OF REDUNDANT ...
    For a binomial distribution, the expected value, E[N(t)] = nF(t). However, replacing E[N(t)] with nF(t) and dividing by n leads to an interesting ...
  62. [62]
    Chapter 14 The binomial distribution | Statistics for Human Genetics
    The binomial distribution allows us to assess the probability of a specified outcome from a series of trials.
  63. [63]
    Probability/Statistics in Inheritance - Genetics
    Sep 14, 2005 · For a simple Mendelian inheritance from two heterozygotic parents, p will = 3/4 (if AA and Aa give dominant phenotype) and q will = 1/4 (aa gives recessive ...
  64. [64]
    4.3 The Binomial Distribution – Significant Statistics
    ” The parameters are n and p: n = number of trials, p = probability of a success on each trial. Since the Binomial counts the number of successes, x, in n ...
  65. [65]
    Probability Playground: The Beta-Binomial Distribution
    The beta-binomial distribution is a discrete distribution generated by choosing the probability p for a binomial(n, p) distribution from a beta distribution ...<|control11|><|separator|>
  66. [66]
    None
    ### Definition and Summary of Beta-Binomial as Marginal of Binomial with Beta Prior on p
  67. [67]
    6: Binary Logistic Regression | STAT 504
    One is that instead of a normal distribution, the logistic regression response has a binomial distribution (can be either "success" or "failure"), and the other ...
  68. [68]
    A multivariate zero-inflated binomial model for the analysis of ... - NIH
    In this paper, a new multivariate zero-inflated binomial (MZIB) distribution is proposed to analyse the correlated proportional data with excessive zeros.
  69. [69]
    Multivariate distributions of correlated binary variables generated by ...
    Mar 5, 2021 · In this paper, we study multivariate binary distributions that are based on D-vine pair-copula models as a superior alternative to these methods.Missing: binomial | Show results with:binomial
  70. [70]
    A comparison of observation-level random effect and Beta-Binomial ...
    Jul 21, 2015 · Beta-binomial models are frequently employed in the ecological literature to model non-independence among probabilities (e.g., Hughes & Madden, ...
  71. [71]
    Phase transition in the Bayesian estimation of the default portfolio
    Feb 11, 2019 · In this paper, we introduce a hierarchical Bayesian estimation method using the beta binomial distribution and consider a multi-year case with a temporal ...
  72. [72]
    [PDF] Moody's Correlated Binomial Default Distribution
    Aug 10, 2004 · Dropping the independence assumption does complicate the calculation of the default probabilities, but it can lead to distributions that have ...<|control11|><|separator|>
  73. [73]
    distributions - Poisson-binomial vs. Beta-binomial - Cross Validated
    Jan 30, 2019 · Of course, by artificially reducing the variance to approach the mean probability value, the beta-binomial will resemble the equivalent poisson ...Beta-Binomial regression or Poisson-Gamma model to account for ...Visually compare binomial and beta-binomial appropriateness in RMore results from stats.stackexchange.com