Slide 11-1 — Discrete Probability Distribution

第11-1页——离散概率分布

Knowledge Points (知识点)

  1. Random Variable(随机变量)
  2. Discrete Probability Distribution(离散概率分布)
  3. Expected Value and Variance(期望值与方差)

🔹 Knowledge Point 1: Random Variable(随机变量)

  • Explanation (解释):
    A random variable is a function that assigns numerical values to the outcomes of a random experiment.
    随机变量是为每个随机试验结果分配一个数值的函数,用于把“事件”转化为“数值”。

  • Example (例子):
    If we toss a coin three times and let X represent the number of heads, then X can take values 0, 1, 2, or 3.
    掷硬币三次,若 X 表示出现正面的次数,则 X 的取值为 0, 1, 2, 3。

  • Extension (拓展):
    Random variables can be classified as:

    • Discrete (离散型): Finite or countable values, e.g., number of customers per day.
    • Continuous (连续型): Infinite values, e.g., transaction amount or time.
      随机变量可分为离散型(可数)与连续型(不可数),二者决定采用的概率模型类型。
  • Summary (总结):
    随机变量是将现实不确定性量化的关键,使概率分析成为可能。


🔹 Knowledge Point 2: Discrete Probability Distribution(离散概率分布)

  • Explanation (解释):
    A discrete probability distribution lists all possible values of a discrete variable and their probabilities.
    离散概率分布列出了所有可能的离散取值及其对应概率。

  • Example (例子):
    The number of daily defective products in a factory may follow a discrete distribution.
    例如工厂每天的不良品数量可视为一个离散概率分布。

  • Extension (拓展):
    In business, discrete distributions are useful for modeling counts — such as service requests, claims, or transactions.
    在商业中,离散分布用于描述“次数类”事件,如客户投诉数、交易次数等。

  • Summary (总结):
    离散概率分布提供了预测事件频率的数学工具,常用于运营和风险分析。


🔹 Knowledge Point 3: Expected Value and Variance(期望与方差)

  • Explanation (解释):
    The expected value (mean) represents the long-run average outcome, and variance measures its spread.
    期望表示随机变量长期平均结果,方差衡量数据的离散程度。

  • Example (例子):
    If the expected daily demand is 100 units with a variance of 25, the company can expect demand to fluctuate around 100 by ±5.
    若每日平均需求为100且方差为25,则需求通常波动在±5范围内。

  • Extension (拓展):
    Expected value aids budgeting and planning, while variance informs risk assessment.
    期望用于预算与规划,方差用于评估波动风险,是投资与库存控制中的关键指标。

  • Summary (总结):
    期望与方差构成统计预测的核心,连接理论与商业实践。


Slide 11-2 — Binomial Probability Distribution (Basics)

第11-2页——二项分布基础

Knowledge Points (知识点)

  1. Definition & Assumptions(定义与假设)
  2. Probability of Success & Failure(成功与失败概率)
  3. Graph Interpretation(图表分析)

🔹 Knowledge Point 1: Definition & Assumptions(定义与假设)

  • Explanation (解释):
    The binomial distribution describes the number of successes in a fixed number (n) of independent trials, where each trial has two outcomes: success or failure.
    二项分布用于描述在 n 次独立重复试验中,成功事件出现的次数。

  • Example (例子):
    A marketing campaign where each customer has a 30% chance of purchase — out of 10 customers, how many will buy?
    一次营销活动中,每位客户有30%的购买概率,在10名客户中可能的购买人数服从二项分布。

  • Extension (拓展):
    This model is widely used in finance (default probability), HR (recruitment success), and production (defect rate).
    二项分布常用于信贷风险、招聘成功率及质量控制分析。

  • Summary (总结):
    二项分布为“固定次数、两种结果”的试验提供概率模型。


🔹 Knowledge Point 2: Probability of Success & Failure(成功与失败概率)

  • Explanation (解释):
    Each trial has probability of success p and failure (1–p), and these remain constant across trials.
    每次试验成功的概率为 p,失败概率为 (1–p),且二者在整个试验中保持不变。

  • Example (例子):
    If p = 0.3, then in 10 trials the most likely number of successes is around 3.
    若 p=0.3,重复10次试验中最可能出现3次成功。

  • Extension (拓展):
    Consistent probability is key to binomial validity; if p changes (e.g., learning effects), results deviate.
    成功概率恒定是二项分布的必要条件,若 p 变化(如学习效应出现),模型将失效。

  • Summary (总结):
    成功与失败的固定概率体现了试验的独立性与稳定性。


🔹 Knowledge Point 3: Graph Interpretation(图表解读)

  • Explanation (解释):
    The chart (n=10, p=0.3) shows the probability of x successes, peaking at x=3 (≈0.2668).
    图中 n=10, p=0.3 的分布峰值出现在 x=3,对应概率约为0.2668。

  • Example (例子):
    In a sales setting, this means out of 10 customer calls, 3 likely lead to sales.
    在销售场景中,这意味着10次电话销售中最可能有3次成交。

  • Extension (拓展):
    The skewed shape indicates low p values cause probabilities to cluster at lower x.
    当 p 较小时,分布右偏;当 p 较大时,分布左偏。企业可据此调整试验策略以提升成功率。

  • Summary (总结):
    图表可视化了“成功次数与概率”的关系,帮助识别最可能结果与波动趋势。


Slide 11-3 — Probability Derivation

第11-3页——二项分布概率推导

Knowledge Points (知识点)

  1. Independence Rule(独立性法则)
  2. One Success in Two Trials(两次试验中一次成功)
  3. Probability Completeness(概率完备性)

🔹 Knowledge Point 1: Independence Rule(独立性法则)

  • Explanation (解释):
    When trials are independent, joint probability equals the product of individual probabilities.
    独立试验的联合概率等于各自概率的乘积:P(A∩B)=P(A)P(B)。

  • Example (例子):
    Tossing two coins: P(two heads) = 0.5 × 0.5 = 0.25.
    投两枚硬币,两次正面概率为 0.25。

  • Extension (拓展):
    Independence ensures no trial affects another — essential for valid binomial modeling.
    独立性保证各试验不受影响,是二项分布成立的前提条件。

  • Summary (总结):
    独立性连接概率的乘法原理,是多次试验计算的核心。


🔹 Knowledge Point 2: One Success in Two Trials(两次试验中一次成功)

  • Explanation (解释):
    For two trials, probability of exactly one success = 2p(1–p).
    若两次试验仅一次成功,概率为 2p(1–p)。

  • Example (例子):
    If p=0.4 → P(1 success) = 2×0.4×0.6 = 0.48.
    当 p=0.4 时,恰好成功一次的概率为 0.48。

  • Extension (拓展):
    This concept extends to n trials via combination coefficients (C(n,x)).
    该逻辑可推广至 n 次试验,通过组合数计算所有可能路径。

  • Summary (总结):
    单次成功概率展示了二项分布的对称结构:成功与失败平衡决定形态。


🔹 Knowledge Point 3: Probability Completeness(概率完备性)

  • Explanation (解释):
    Total probability of all outcomes equals 1.
    所有结果的概率之和为1,保证分布完备性。

  • Example (例子):
    For two trials: P(0)+P(1)+P(2)= (1–p)² + 2p(1–p) + p² = 1.
    两次试验中所有情形概率之和恒为1。

  • Extension (拓展):
    Ensures every possible event accounted — fundamental for risk modeling.
    概率完备性保证所有可能事件均被涵盖,是风险分析与控制模型的基础。

  • Summary (总结):
    概率完备性使模型可封闭地反映所有现实可能性。


Slide 11-4 — Binomial Probability Formula

第11-4页——二项分布公式

Knowledge Points (知识点)

  1. Formula Derivation(公式推导)
  2. Factorial Concept(阶乘概念)
  3. Statistical Application(统计应用)

🔹 Knowledge Point 1: Formula Derivation(公式推导)

  • Explanation (解释):
    The general formula for x successes in n trials:
    f(x) = n! / [x!(n–x)!] × p^x × (1–p)^(n–x)
    该公式结合了组合数学与概率原理。

  • Example (例子):
    n=2, x=1, p=0.3 → f(1)=2×0.3×0.7=0.42。
    表示两次试验中恰好一次成功的概率为42%。

  • Extension (拓展):
    Each term combines “ways to succeed” and “probability of each path.”
    公式中的组合项表示成功路径数,幂项反映每条路径的成功概率。

  • Summary (总结):
    二项分布公式揭示了事件概率来源:组合数量 × 成功概率乘积。


🔹 Knowledge Point 2: Factorial Concept(阶乘概念)

  • Explanation (解释):
    n! = n×(n–1)×(n–2)…×1 represents total permutations.
    阶乘表示从 n 个对象中全部排列的数量。

  • Example (例子):
    5! = 120 → five objects have 120 possible orderings.
    若 n=5,则有120种排列可能。

  • Extension (拓展):
    Factorials simplify probability computation for combinational events in quality testing.
    阶乘在计算组合事件概率时简化了复杂的路径计数,如产品抽检中的合格组合数。

  • Summary (总结):
    阶乘是概率计算的核心工具,用于确定事件出现的组合方式。


🔹 Knowledge Point 3: Statistical Application(统计应用)

  • Explanation (解释):
    From the binomial formula, expected value and variance can be derived:
    E[X] = np, Var[X] = np(1–p)
    由二项分布公式可推导期望与方差,揭示分布集中与离散程度。

  • Example (例子):
    For n=10, p=0.3 → E[X]=3, Var[X]=2.1。
    表示10次试验中平均成功3次,波动约为√2.1 ≈ 1.45。

  • Extension (拓展):
    Used in planning (expected sales calls success), forecasting (machine failure), and decision optimization.
    二项分布结果用于预测销售成功率、设备故障概率及优化决策方案。

  • Summary (总结):
    二项分布的数学性质支持企业从“概率”转向“预测”,实现量化决策。


Slide 11-5 — Example: Evans Electronics (Binomial Application)

第11-5页——案例分析:Evans Electronics(二项分布应用)

Knowledge Points (知识点)

  1. Problem Context(问题背景)
  2. Probability Setup(概率设定)
  3. Decision Implications(决策意义)

🔹 Knowledge Point 1: Problem Context(问题背景)

  • Explanation (解释):
    Evans Electronics faces a 10% annual turnover among hourly employees. Management randomly selects 3 employees and asks: “What is the probability that exactly one of them will leave next year?”
    Evans Electronics 每年时薪员工离职率为 10%。管理层随机抽取 3 名员工,求恰有 1 人离职的概率。

  • Example (例子):
    This is a classic binomial experiment, where:
    n = 3, p = 0.1, x = 1.
    每次试验代表“员工是否离职”,三次独立试验中求一次成功(离职)。

  • Extension (拓展):
    Such analysis helps HR departments forecast turnover and plan replacements or training costs.
    此类概率分析帮助人力资源部门预测离职人数,以便进行培训与替补计划。

  • Summary (总结):
    本页设定了实际商业场景,展示了二项分布在员工流动分析中的应用。


🔹 Knowledge Point 2: Probability Setup(概率设定)

  • Explanation (解释):
    Each employee has a 0.1 probability of leaving and 0.9 of staying, independently.
    每名员工离职概率为 0.1,留下的概率为 0.9,三名员工间相互独立。

  • Example (例子):
    For Mike, Alice, and Barbara:

    • Mike leaves, others stay → (0.1)(0.9)(0.9)=0.081
    • Alice leaves, others stay → (0.9)(0.1)(0.9)=0.081
    • Barbara leaves, others stay → (0.9)(0.9)(0.1)=0.081
      Total P(1 leave)=0.081×3=0.243。
  • Extension (拓展):
    HR analytics can apply this method to model multiple risk events — for example, “probability that one team member resigns.”
    HR分析中可利用此模型计算团队中单人离职或缺勤的风险。

  • Summary (总结):
    通过乘法原理与组合数计算,可快速得出“恰好一次成功”的概率。


🔹 Knowledge Point 3: Decision Implications(决策意义)

  • Explanation (解释):
    The result P(1 leave)=0.243 indicates a 24.3% chance of exactly one departure.
    结果显示,恰有一人离职的概率为 24.3%。

  • Example (例子):
    This means that in a group of three, about 1 in 4 such selections will have exactly one employee leave.
    约每四组三人中,就有一组会出现一人离职。

  • Extension (拓展):
    Management can use such models for staffing risk estimation and retention budgeting.
    管理层可据此估算人力流失风险并制定保留预算。

  • Summary (总结):
    二项分布结果可量化人力资源风险,为决策提供数据支撑。


Slide 11-6 — Evans Electronics (Probability Calculation Table)

第11-6页——Evans Electronics 概率表分析

Knowledge Points (知识点)

  1. Probability Structure(概率结构)
  2. Formula Calculation(公式计算)
  3. Visualization Analysis(图表分析)

🔹 Knowledge Point 1: Probability Structure(概率结构)

  • Explanation (解释):
    The probability for each employee scenario follows the pattern (0.1)(0.9)(0.9)=0.081. There are 3 such combinations for one success.
    每个“恰有一人离职”的情形概率均为0.081,总共有3种可能。

  • Example (例子):

    • Mike leaves → 0.081
    • Alice leaves → 0.081
    • Barbara leaves → 0.081
      Total = 0.243.
  • Extension (拓展):
    This pattern generalizes to n trials: number of arrangements × individual probability paths = total probability.
    该模式可推广至 n 次试验,即“排列组合 × 各路径概率”得总概率。

  • Summary (总结):
    通过表格清晰展示了二项分布概率的逻辑来源。


🔹 Knowledge Point 2: Formula Calculation(公式计算)

  • Explanation (解释):
    The binomial formula:
    f(x) = n! / [x!(n–x)!] × p^x × (1–p)^(n–x).
    Here n=3, x=1, p=0.1 → f(1)=3×0.1×0.81=0.243。

  • Example (例子):
    Using factorials: 3!/[1!(3–1)!]=3 ways for one to leave.
    阶乘项表示三种可能的离职组合。

  • Extension (拓展):
    In HR data analytics, this formula supports predictive retention modeling (probability of future turnover).
    在人力分析中,该公式可用于预测未来离职人数分布。

  • Summary (总结):
    二项分布公式将“路径组合”数学化,为概率计算提供通用框架。


🔹 Knowledge Point 3: Visualization Analysis(图表分析)

  • Explanation (解释):
    The table visually pairs each employee scenario with its corresponding probability.
    表格将每个员工离职/留任组合与其概率相匹配,体现了结构化概率计算。

  • Example (例子):
    Each row shows one “success” case; combined, they form the event P(x=1).
    每行代表一种成功一次的情况,三行构成总体概率。

  • Extension (拓展):
    Visualization like this enhances clarity in risk communication and management reporting.
    此类图表帮助管理者直观看到各事件概率构成,便于汇报与解释。

  • Summary (总结):
    表格与公式结合验证了二项分布计算的准确性与透明性。


Slide 11-7 — Evans Electronics (Expected Value & Variance)

第11-7页——Evans Electronics 的期望值与方差

Knowledge Points (知识点)

  1. Expected Value(期望值)
  2. Variance and Standard Deviation(方差与标准差)
  3. Excel Function Application(Excel函数应用)

🔹 Knowledge Point 1: Expected Value(期望值)

  • Explanation (解释):
    The expected number of employees leaving is given by E[X] = np = 3×0.1=0.3。
    期望离职人数为 0.3,表示平均每三名员工中有 0.3 名离职。

  • Example (例子):
    If the firm has 300 hourly employees, expected departures = 0.3/3×300=30 per year.
    若公司有300名员工,则预计每年约30人离职。

  • Extension (拓展):
    HR can use expected turnover to forecast hiring and replacement needs.
    期望值用于预测招聘需求及人力补充计划。

  • Summary (总结):
    期望值反映离职风险的平均水平,是人力成本预测的重要指标。


🔹 Knowledge Point 2: Variance and Standard Deviation(方差与标准差)

  • Explanation (解释):
    Variance Var(X)=np(1–p)=3×0.1×0.9=0.27;
    Standard deviation σ=√0.27≈0.52。
    方差衡量离职人数的波动范围,标准差表示平均偏离度。

  • Example (例子):
    A σ of 0.52 implies that in most 3-employee groups, turnover fluctuates between 0 and 1.
    标准差 0.52 表示大部分三人组离职人数介于 0 至 1 之间。

  • Extension (拓展):
    Variance helps gauge the stability of retention rates — smaller variance implies consistent HR performance.
    方差越小,说明离职率越稳定,人力管理越有效。

  • Summary (总结):
    方差与标准差体现风险波动水平,是衡量人事稳定度的重要工具。


🔹 Knowledge Point 3: Excel Function Application(Excel函数应用)

  • Explanation (解释):
    Binomial probabilities can be computed using Excel:

    • Non-cumulative: =BINOM.DIST(1,10,0.1,FALSE)
    • Cumulative: =BINOM.DIST(1,10,0.1,TRUE)
      Excel 可直接求出单次或累计成功概率。
  • Example (例子):
    For n=10, p=0.1, x=1 → f(1)=0.3874.
    在10次试验中恰有一次成功的概率约为0.3874。

  • Extension (拓展):
    Excel enables fast simulation and sensitivity testing for HR or marketing analytics.
    Excel 函数使概率模拟与敏感性分析更加便捷,适合企业批量计算。

  • Summary (总结):
    运用Excel可在实际工作中快速验证概率模型,提高决策效率。


Slide 11-8 — Poisson Probability Distribution

第11-8页——泊松分布(Poisson Distribution)

Knowledge Points (知识点)

  1. Concept and Usage(概念与应用)
  2. Key Assumptions(主要假设)
  3. Real-life Example(现实案例)

🔹 Knowledge Point 1: Concept and Usage(概念与应用)

  • Explanation (解释):
    The Poisson distribution estimates the number of occurrences in a fixed time or space interval.
    泊松分布用于估计固定时间或空间区间内事件发生的次数。

  • Example (例子):
    Number of customers arriving at a store between 2–3 pm.
    例如下午2点至3点之间进店顾客人数。

  • Extension (拓展):
    Often used for modeling rare or random events — defects, accidents, or system failures.
    常用于描述稀有事件,如设备故障、事故发生率、网络请求数等。

  • Summary (总结):
    泊松分布在“随机到达”类事件建模中应用广泛。


🔹 Knowledge Point 2: Key Assumptions(主要假设)

  • Explanation (解释):

    • Events occur independently;
    • Average rate (λ) is constant;
    • Two events cannot occur simultaneously.
      假设事件独立、平均发生率 λ 恒定,且同时事件不会出现。
  • Example (例子):
    Calls arriving at a help desk at an average of 10 per hour follow Poisson(λ=10).
    每小时平均10个客户来电,来电次数服从参数 λ=10 的泊松分布。

  • Extension (拓展):
    In queue management, Poisson helps forecast service demand and optimize staffing.
    在排队与客服管理中,泊松分布可预测服务需求并优化人力分配。

  • Summary (总结):
    该分布适用于独立稀疏事件,有助于改进运营效率。


🔹 Knowledge Point 3: Real-life Example(现实案例)

  • Explanation (解释):
    Example: the number of patients arriving in an emergency room between 10–11 pm.
    例:晚上10至11点进入急诊室的病人数。

  • Example (例子):
    If the average λ=5, then P(x=3)= (5³e⁻⁵)/3! = 0.1404.
    若平均λ=5,则1小时内恰有3人到来的概率为0.1404。

  • Extension (拓展):
    Hospitals use Poisson modeling to plan staff shifts and capacity for emergency units.
    医院通过泊松模型预测病人流量并规划急诊人力。

  • Summary (总结):
    泊松模型帮助机构量化随机事件,提升资源分配与风险管理水平。


Slide 11-9 — Poisson Probability Distribution (Formula & Characteristics)

第11-9页——泊松概率分布(公式与特征)

Knowledge Points (知识点)

  1. Definition and Properties(定义与性质)
  2. Probability Formula(概率公式)
  3. Excel Application(Excel函数应用)

🔹 Knowledge Point 1: Definition and Properties(定义与性质)

  • Explanation (解释):
    The Poisson distribution models the number of occurrences of an event within a fixed interval of time or space when these events happen independently and at a constant mean rate (μ).
    泊松分布用于描述在固定时间或空间区间内,事件以恒定速率独立发生的次数分布。

  • Example (例子):
    Number of incoming customer calls per hour in a call center.
    呼叫中心每小时接到的电话数量。

  • Extension (拓展):
    Poisson is used for rare or count-type events such as defects, arrivals, or accidents.
    泊松分布常用于建模稀有或“计数类”事件,如产品缺陷数、顾客到达数、交通事故数等。

  • Summary (总结):
    泊松分布关注“事件发生的次数”,是离散型分布中最常见的计数模型之一。


🔹 Knowledge Point 2: Probability Formula(概率公式)

  • Explanation (解释):
    The probability of observing x occurrences:
    f(x) = (μ^x * e^(-μ)) / x!
    where μ is the average rate, e = 2.71828.
    泊松分布的概率函数:f(x) = (μ^x * e^(-μ)) / x!,其中 μ 为平均发生率,e 为自然常数。

  • Example (例子):
    For μ = 10, x = 1, f(1) = (10¹ * e⁻¹⁰) / 1! = 0.00045。
    若平均每小时10次事件,恰有1次发生的概率为0.00045。

  • Extension (拓展):
    In Poisson, both mean and variance are equal: μ = σ².
    This means higher μ increases both the average and variability of occurrences.
    泊松分布具有 μ = σ² 的特征,表示均值越高,波动性也越大。

  • Summary (总结):
    泊松分布的数学形式简洁,能精确描述独立事件的出现频率。


🔹 Knowledge Point 3: Excel Application(Excel函数应用)

  • Explanation (解释):
    In Excel, probabilities can be calculated using:

    • Non-cumulative: =POISSON.DIST(x, μ, FALSE)
    • Cumulative: =POISSON.DIST(x, μ, TRUE)
      Excel 函数用于计算泊松分布的单点与累计概率。
  • Example (例子):
    If μ = 10, x = 1 → =POISSON.DIST(1,10,FALSE) = 0.00045.
    若平均值为10,恰好一次的概率即为0.00045。

  • Extension (拓展):
    Useful in HR (absence modeling), logistics (arrival forecasting), or IT (server error counts).
    在人力资源(缺勤建模)、物流(到达预测)、信息系统(服务器错误计数)中广泛应用。

  • Summary (总结):
    Excel 函数使泊松分布计算自动化,为商业分析提供快捷工具。


Slide 11-10 — Poisson Distribution (Shape & Range)

第11-10页——泊松分布(形态与范围)

Knowledge Points (知识点)

  1. Range of X(取值范围)
  2. Distribution Shape(分布形态)
  3. Graph Interpretation(图形解读)

🔹 Knowledge Point 1: Range of X(取值范围)

  • Explanation (解释):
    Theoretically, x can take any non-negative integer (0, 1, 2, 3, … ∞).
    理论上,x 可取任意非负整数(0, 1, 2, 3,… ∞)。

  • Example (例子):
    Counting the number of emails received per hour can result in any integer value.
    每小时收到的邮件数可能为0、1、2、3等任意整数。

  • Extension (拓展):
    Practically, when x becomes very large, f(x) approaches zero and becomes negligible.
    实际上,当 x 过大时,f(x) 近似为0,可忽略不计。

  • Summary (总结):
    泊松分布虽然理论上无限延伸,但实际应用中仅有限值范围有意义。


🔹 Knowledge Point 2: Distribution Shape(分布形态)

  • Explanation (解释):
    The shape depends on μ:

    • Small μ → Right-skewed (rare events).
    • Large μ → More symmetric (approaches normal distribution).
      分布形态由 μ 决定,μ 小则偏右,μ 大则趋向正态分布。
  • Example (例子):
    For μ = 3, the probability peaks at x = 2, 3 (≈0.2240 each).
    当 μ=3 时,峰值出现在 x=2 或 3,概率均为0.2240。

  • Extension (拓展):
    Managers use this property to interpret frequency charts — e.g., daily complaint patterns.
    管理者可据此分析频率分布,如每日投诉数量模式。

  • Summary (总结):
    分布形态揭示了事件发生频率的集中趋势与变化范围。


🔹 Knowledge Point 3: Graph Interpretation(图形分析)

  • Explanation (解释):
    The graph shows F(x) for μ=3 declining rapidly after x=5.
    图表显示 μ=3 的泊松分布在 x>5 后快速下降。

  • Example (例子):
    Most likely outcomes (x=2 or 3) have the highest probability (≈0.2240).
    最可能结果为2或3次发生。

  • Extension (拓展):
    In business analytics, this helps estimate optimal resource allocation where demand is random.
    在商业分析中,可用于预测随机需求下的最优资源配置(如客服人数)。

  • Summary (总结):
    图形揭示了“事件次数越多→概率越小”的规律,体现随机事件的稀疏性。


Slide 11-11 — Example: Mercy Hospital

第11-11页——案例:Mercy Hospital

Knowledge Points (知识点)

  1. Case Context(案例背景)
  2. Probability Calculation(概率计算)
  3. Real-world Application(实际意义)

🔹 Knowledge Point 1: Case Context(案例背景)

  • Explanation (解释):
    Patients arrive at the emergency room at an average rate of 6 per hour.
    We want to find the probability that exactly 4 arrive in 30 minutes.
    Mercy Hospital 急诊室每小时平均有6名病人到达,求30分钟内恰有4人到达的概率。

  • Example (例子):
    Since 30 minutes = 0.5 hour, μ = 6×0.5 = 3.
    由于30分钟是半小时,因此 μ=3。

  • Extension (拓展):
    This scenario demonstrates time scaling in Poisson processes — adjusting μ for shorter intervals.
    该例展示泊松过程的时间比例特性:可按时间长度调整 μ。

  • Summary (总结):
    案例体现泊松模型在医疗急诊管理中的实用性。


🔹 Knowledge Point 2: Probability Calculation(概率计算)

  • Explanation (解释):
    f(4) = (3⁴ * e⁻³) / 4! = 0.1680
    表示半小时内恰有4人到达的概率为0.168。

  • Example (例子):
    计算步骤:
    μ⁴ = 81,e⁻³ = 0.0498,4! = 24 → f(4)=81×0.0498/24=0.168。

  • Extension (拓展):
    Hospitals can use this to predict patient arrival rates and optimize staffing.
    医院可据此预测急诊流量,合理安排医生和护士班次。

  • Summary (总结):
    泊松分布帮助医疗机构用数据支撑人力调度与资源分配。


🔹 Knowledge Point 3: Real-world Application(实际意义)

  • Explanation (解释):
    Poisson helps model unpredictable yet measurable random arrivals.
    泊松模型适用于预测难以精确控制但有统计规律的事件到达。

  • Example (例子):
    Similar models are used in banking (loan requests), call centers, and logistics deliveries.
    银行贷款申请、客服来电、物流投递等皆可用泊松分布建模。

  • Extension (拓展):
    Integrating Poisson models with time-series tools improves operational forecasting accuracy.
    结合时间序列分析可进一步提升运营预测精度。

  • Summary (总结):
    案例揭示了泊松模型的跨行业适用性和预测能力。


Slide 11-12 — Example: Baby Heart Disease

第11-12页——案例:新生儿心脏病

Knowledge Points (知识点)

  1. Medical Probability Context(医学概率背景)
  2. Calculation Process(计算过程)
  3. Interpretation & Practical Use(结果解读与实际意义)

🔹 Knowledge Point 1: Medical Probability Context(医学背景)

  • Explanation (解释):
    Among newborns, 8 out of 1000 have heart disease. Among 120 infants, what’s the probability that 4 have it?
    在1000名新生儿中有8名患心脏病,求120名婴儿中有4名患病的概率。

  • Example (例子):
    u = 0.008×120 = 0.96, x = 4。
    μ 表示120个样本中平均病例数为 0.96。

  • Extension (拓展):
    This type of data is common in epidemiology and health risk modeling.
    此类分析常见于流行病学与公共卫生风险建模。

  • Summary (总结):
    医学案例展示泊松分布在疾病概率预测中的作用。


🔹 Knowledge Point 2: Calculation Process(计算过程)

  • Explanation (解释):
    For x=4: P(x)= (0.96⁴ * e⁻⁰․⁹⁶) / 4! = 0.014
    For x=1: P(x)= (0.96¹ * e⁻⁰․⁹⁶) / 1! = 0.3676
    分别计算4人与1人患病的概率。

  • Example (例子):
    使用 e=2.71828 代入计算,可得结果:

    • 4人患病概率 1.4%
    • 1人患病概率 36.8%
  • Extension (拓展):
    Low probabilities reflect rarity — supports medical resource prioritization.
    低概率结果反映疾病罕见,有助于公共卫生资源的合理分配。

  • Summary (总结):
    计算揭示罕见疾病分布的泊松特性:期望低、方差小、事件稀少。


🔹 Knowledge Point 3: Interpretation & Practical Use(结果解读与实际意义)

  • Explanation (解释):
    Probabilities can be validated using Excel’s =POISSON.DIST(4,0.96,FALSE) and =POISSON.DIST(1,0.96,FALSE).
    可用Excel函数验证计算结果,简化实际应用。

  • Example (例子):
    Epidemiologists may use this approach to estimate outbreak likelihood.
    流行病学家通过该模型估算疫情爆发或疾病出现频率。

  • Extension (拓展):
    Combining Poisson models with demographic data enhances preventive healthcare planning.
    将泊松模型与人口数据结合,可制定针对性的公共健康预防措施。

  • Summary (总结):
    本案例说明泊松分布在医疗与风险预测领域的强大解释力。


Slide 11-13 — Binomial vs. Poisson

第11-13页——二项分布与泊松分布对比

Knowledge Points (知识点)

  1. Relationship Between Binomial and Poisson(两者关系)
  2. Mean and Variance Comparison(期望与方差对比)
  3. Graph Interpretation(图形分析)

🔹 Knowledge Point 1: Relationship Between Binomial and Poisson(两者关系)

  • Explanation (解释):
    The Poisson distribution can be used as an approximation to the binomial distribution when n is large and p is small, with μ = np.
    当试验次数 n 很大且成功概率 p 很小时,泊松分布可近似代替二项分布,其中 μ = np。

  • Example (例子):
    For a process with n=10, p=0.3 → μ = np = 3.
    比如 n=10, p=0.3 时,泊松分布参数 μ=3,可近似描述同样的事件模式。

  • Extension (拓展):
    This approximation simplifies calculations for rare events, such as defect counts in manufacturing or server errors in IT systems.
    该近似常用于稀有事件概率计算,如制造缺陷数量或网络服务器错误次数。

  • Summary (总结):
    当事件概率小而试验次数多时,泊松分布是二项分布的极限形式。


🔹 Knowledge Point 2: Mean and Variance Comparison(期望与方差对比)

  • Explanation (解释):

    • Binomial: E(X)=np, Var(X)=np(1–p)
    • Poisson: E(X)=μ, Var(X)=μ
      二项分布与泊松分布的核心区别在于方差公式:泊松分布方差与均值相等。
  • Example (例子):
    For n=10, p=0.3 → E=3, Var=2.1 (Binomial);
    For Poisson with μ=3 → E=3, Var=3.
    二项分布的波动性略小于泊松分布。

  • Extension (拓展):
    This difference affects prediction accuracy — binomial is more precise for small n, Poisson is better for large-scale rare events.
    方差差异影响预测精度:n 较小时应选二项分布,事件稀疏时适用泊松分布。

  • Summary (总结):
    泊松分布的方差较大,反映了事件更高的不确定性与波动性。


🔹 Knowledge Point 3: Graph Interpretation(图形分析)

  • Explanation (解释):
    The graph compares Binomial (n=10, p=0.3) and Poisson (μ=3).
    Both show similar shapes — unimodal and right-skewed — with Poisson slightly smoother.
    图中两条曲线均呈右偏单峰形态,泊松分布曲线更平滑。

  • Example (例子):
    Peak probability occurs near X=3 for both distributions.
    两者峰值均出现在 X=3 附近。

  • Extension (拓展):
    When designing reliability systems, using Poisson instead of Binomial reduces computational effort with minimal accuracy loss.
    在可靠性分析中,泊松分布能在保证精度的前提下显著简化计算。

  • Summary (总结):
    图形表明二项分布与泊松分布在稀有事件建模中高度相似,泊松为更简便替代模型。


Slide 11-14 — Binomial Distribution in Cumulative Probability

第11-14页——二项分布的累积概率

Knowledge Points (知识点)

  1. Definition of Cumulative Probability(累积概率定义)
  2. Graph Interpretation(图形解读)
  3. Practical Application(实际应用)

🔹 Knowledge Point 1: Definition of Cumulative Probability(累积概率定义)

  • Explanation (解释):
    Cumulative probability represents the total probability that a variable takes a value less than or equal to x:
    P(X ≤ x) = f(0) + f(1) + … + f(x)。
    累积概率表示随机变量取值不超过某一特定值的总概率。

  • Example (例子):
    For n=15, p=0.5, cumulative probability at X=8 equals the total probability of 0–8 successes.
    当 n=15, p=0.5 时,X=8 的累计概率即从0到8次成功的总和。

  • Extension (拓展):
    Cumulative probability helps estimate thresholds, e.g., “What’s the probability of ≤ 3 defects in a batch?”
    累积概率常用于确定控制阈值,如“一批产品中缺陷数不超过3个的概率”。

  • Summary (总结):
    累积概率是评估总体事件分布区间的重要工具。


🔹 Knowledge Point 2: Graph Interpretation(图形解读)

  • Explanation (解释):
    The blue bars show individual binomial probabilities;
    the orange curve represents the cumulative total approaching 1.
    蓝柱代表各取值的单点概率,橙线代表累计概率随 X 增大逐渐逼近1。

  • Example (例子):
    For Binomial (n=15, p=0.5), cumulative probability reaches ≈1 near X=15.
    在 n=15, p=0.5 情况下,当 X 接近15时,累计概率趋近1。

  • Extension (拓展):
    In finance, cumulative distributions are used to determine value-at-risk (VaR) and credit default likelihoods.
    在金融中,累积分布用于计算风险价值 (VaR) 或信用违约概率。

  • Summary (总结):
    图形揭示累积概率随事件次数上升而趋近完整分布(总概率=1)。


🔹 Knowledge Point 3: Practical Application(实际应用)

  • Explanation (解释):
    Excel provides built-in cumulative functions:
    =BINOM.DIST(x, n, p, TRUE) gives P(X ≤ x).
    Excel 可通过 =BINOM.DIST(x,n,p,TRUE) 快速求得累计概率。

  • Example (例子):
    If n=15, p=0.5, x=8 → =BINOM.DIST(8,15,0.5,TRUE) ≈ 0.566。
    表示成功次数不超过8的概率为约56.6%。

  • Extension (拓展):
    Cumulative probability helps decision-makers set safety limits or acceptance thresholds (e.g., quality inspection).
    在决策中,累计概率用于确定“可接受范围”或“安全上限”。

  • Summary (总结):
    累积概率函数结合 Excel 使用,使统计推断更直观与高效。