所屬科目:研究所、轉學考(插大)◆統計學
一、(10%) Consider the c.d.f. given bySketch f(x) and F(x).
二、(10%) With X₁ denoting the amount of gasoline stocked in a bulk tank at the beginning of a week and X₂ the amount sold during the week, Y = X₁ - X₂ represents the amount left over at the end of the week. Find the mean and variance of Y if the joint function of (X₁, X₂) is given byf(x₁, x₂) = 3x₁ 0 ≤ x₂ ≤ x₁ ≤ 1 = 0 elsewhere
(1) the two components operate in series (5%)
(2) the two components operate in parallel (5%)
四、(10%) A taxi is traveling back and forth along a linear route of k miles for servicing customers who call for a ride. Customer calls may occur randomly at any location on the route. Suppose that the taxi will respond to and go picking the customer immediately when it receives his/her call. Find the average moving distance (in miles) of the taxi from where it answered a call to the customer?
(1) Find the probability that exactly two of the three plants process more than 4 tons on a given day? (5%)
(2) For a particular plant, how much raw sugar should be stocked for that plant each day so that the chance of running out of product is only 0.05? (5%)
六、(10%) 有一呈波松分配(Poisson Distribution)之母體,今欲估計其平均數(Mean)µ,合理規劃之平均值(µ)為 1.5。在 10%之精度要求(即µ±0.1µ)及 95%信賴區間之水準要求下,試問所需之樣本數大小為何?(註:在波松分配下,σ² = μ)
(1) 試問上述之作法為觀察研究?或是實驗研究?(2%)
(2) 請構建 μA-μB 之 95%信賴區間,並解釋其意義。(5%)
(3) 請檢定 A、B 兩款購物指南之得分平均數是否相等?當μA-μB =0 時、請控制 α = 0.05。(5%)
(4) 第(3)項中之檢定 p 值為何?(3%)
八、(10%) 有一研究為調查屠宰用豬從高雄縣運至台北市之失重情況,茲有如下之兩種方法進行此研究:(方法一) 在運送出發前,從所運送之豬隻中隨機抽取 n 頭豬分別稱其重量,並求其平均體重 及變異數 SIB;等到運抵台北時,再一次從所運送之豬隻中隨機抽取 n 頭豬分別稱其重量,並求其平均體重 及變異數 S1A。利用作為μ1B-μ1A 之估計值並進行檢定。(方法二) 在運送出發前,從所運送之豬隻中隨機抽取 n 頭豬,對被抽取之每一頭豬給以之編號,並分別稱其重量 XiB;等到運抵台北時,再一次對運送前已被抽取之每一頭豬稱其重量 XiA,每頭豬兩次所稱之體重差異 D=XiB-XiA 即代表運送過程所造成之失重。最後則以被抽取之 n 頭豬之失重平均值()及變異數(SD)進行檢定。試問上述兩種作法何者之估計精度較高?請說明其理由,並提出必要之數學式作佐證。
(1)試說明以最小平方法校估上述模式時,其參數估計值有何特性?(4%)
(2)如果以最大概似估計法(Maximum Likelihood Estimation; MLE)校估上述模式時,其參數估計值又有何特性?(4%)
(3)在多元迴歸模式之變數篩選過程中,R²值愈高是否表示該模式愈佳?為什麼?(4%)
(4)當模式內之說明變數間出現很高之相關性時,模式之校估結果可能出現那些現象?(3%)