Scribd Logo
Upload

Welcome to Scribd!

  • 45 views

    Assignment Y1: (I) Median

    Uploaded by

    Bakari Hamisi
    This document contains R code for analyzing and simulating various probability distributions. Question 1 calculates the median and binomial probabilities for a binomial distribution. Question 2 simulates an exponential distribution and analyzes its properties. Question 3 analyzes confidence intervals for the binomial distribution. Question 4 analyzes an exponential distribution, simulates sample means from an exponential, compares to a gamma distribution, and performs a QQ plot comparing simulated gamma and sample means.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    Assignment Y1: (I) Median

    Uploaded by

    Bakari Hamisi
    45 views8 pages
    This document contains R code for analyzing and simulating various probability distributions. Question 1 calculates the median and binomial probabilities for a binomial distribution. Question 2 simulates an exponential distribution and analyzes its properties. Question 3 analyzes confidence intervals for the binomial distribution. Question 4 analyzes an exponential distribution, simulates sample means from an exponential, compares to a gamma distribution, and performs a QQ plot comparing simulated gamma and sample means.

    Original Title

    CS1_Y1Assignment

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document contains R code for analyzing and simulating various probability distributions. Question 1 calculates the median and binomial probabilities for a binomial distribution. Question 2 simulates an exponential distribution and analyzes its properties. Question 3 analyzes confidence intervals for the binomial distribution. Question 4 analyzes an exponential distribution, simulates sample means from an exponential, compares to a gamma distribution, and performs a QQ plot comparing simulated gamma and sample means.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    45 views8 pages

    Assignment Y1: (I) Median

    Uploaded by

    Bakari Hamisi
    This document contains R code for analyzing and simulating various probability distributions. Question 1 calculates the median and binomial probabilities for a binomial distribution. Question 2 simulates an exponential distribution and analyzes its properties. Question 3 analyzes confidence intervals for the binomial distribution. Question 4 analyzes an exponential distribution, simulates sample means from an exponential, compares to a gamma distribution, and performs a QQ plot comparing simulated gamma and sample means.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    You are on page 1of 8

    Assignment Y1

    Question 1

    (i) Median
    n<-1000
    p<-0.05
    qbinom(0.5,n,p)

    ## [1] 50

    (ii) Probability that the number of deaths D lies between 45 and 59


    inclusive
    #(a) exactly
    pbinom(59,n,p)-pbinom(45,n,p)

    ## [1] 0.6523035

    #(b) using a Poisson approximation


    # Since Bin(n,p) is approximately Pois(np), we have a Poisson Distribution
    with mean 50
    ppois(59,n*p)-ppois(45,n*p)

    ## [1] 0.6408685

    #(c) using a normal approximation


    #Since Bin(n,p) is approximately N(np, npq), we have
    pnorm(59,50,sqrt(47.5)) - pnorm(45, 50, sqrt(47.5))

    ## [1] 0.6701195
    Question 2

    (i) (a) Simulated values


    set.seed(13)
    u<-runif(1000,0,1)
    u

    (i) (b)
    Rearranging u=F ( x ) to get x=F−1 (u)
    1 1
    u=1− becomes x= −1
    1+ x 1−u
    x<- (1-u)^(-1)-1
    x

    (ii) (a)
    plot(density(x),xlim=c(0,200), main="Empirical PDF of the simulations",
    xlab="x", col="blue")
    (ii) (a)
    mean(x)

    ## [1] 14.54969

    sd(x)

    ## [1] 199.7557

    skew<-sum((x-mean(x))^3)/length(x)

    skew/(sd(x)^3)

    ## [1] 27.72327

    The huge standard deviation for such a small mean indicates that we have a very long tail
    (as the values must be greater than zero).
    This is confirmed by the very large positive coefficient of skewness.
    Question 3

    (i) Width of as exact 95% confidence interval


    n<-20
    x<-10
    test<-binom.test(x,n,conf=0.95)
    test$conf[2]-test$conf[1]

    ## [1] 0.4560843

    Hence the width is approximately45.6.

    (ii) Showing that the greatest width occurs when there are 10
    successes
    width<-rep(0,20)
    for (i in 1:20)
    {test<-binom.test(i,20); width[i]<-test$conf[2]-test$conf[1]}

    width

    ## [1] 0.2474677 0.3046342 0.3468559 0.3792800 0.4044744 0.4238576 0.4382793


    ## [8] 0.4482674 0.4541440 0.4560843 0.4541440 0.4482674 0.4382793 0.4238576
    ## [15] 0.4044744 0.3792800 0.3468559 0.3046342 0.2474677 0.1684335

    max(width)

    ## [1] 0.4560843

    By examining the widths, we see that the greatest width (0.4560843) occurs for 10
    successes.
    Question 4

    (i) 99% Confidence interval for the exponential parameter


    data<-c(14, 4, 3, 2, 3, 1, 5, 10, 4, 23)
    n<-length(data)
    bm<-rep(0,1000)

    set.seed(17)
    for (i in 1:1000)
    {y<-sample(data,replace=TRUE);bm[i]<-mean(y)}
    ci<-quantile(bm,c(0.005,0.995))

    Therefore a 99 % confidence interval for λ is given by:


    1/ci

    ## 0.5% 99.5%
    ## 0.37043897 0.07517949

    This is (0.07518, 0.3704).

    (ii) (a) Simulating 1000 sample means from Exp(0.145)


    l<-0.145
    xbar<-rep(0,1000)
    set.seed(19)
    for (i in 1:1000)
    {x<-rexp(10,l);xbar[i]<-mean(x)}

    (ii) (b) Histogram


    hist(xbar, prob=TRUE, main="Densities of the sample means from Exp(0.145)",
    xlab="Sample means")
    (ii) (c) Probability that empirical mean is less than 5
    length(xbar[xbar<5])/length(xbar)

    ## [1] 0.183

    This gives an answer of 0.183.

    (iii) Gamma
    Using X ∼ Gamma(n , nλ), we have X ∼ Gamma(10,1.45)
    xvals<-seq(0,20,by=0.01)
    hist(xbar, prob=TRUE, main="Densities of the sample means from Exp(0.145)",
    xlab="Sample means")
    lines(xvals, dgamma(xvals, 10, 1.45), type="l", col="blue")
    The histogram has
    a similar shape to the gamma distribution.

    (iv) Exact probability


    pgamma(5,10,1.45)

    ## [1] 0.1957324

    This gives an answer of 0.19573. There is about a 7% difference in the answers so they’re
    not that close. # (v) (a) Simulation
    set.seed(21)
    sim<-rgamma(1000,10,1.45)

    (v) (b) QQ plot


    qqplot(sim, xbar, xlab="Gamma quantiles", ylab="Sample mean quantiles")
    abline(0,1,col="red", lty=2, lwd=2)
    The fit appears fairly good in the middle. The lower end sample means are slightly higher
    than expected - so we have a lighter lower tail. The upper end sample means are much
    lower than expected - so we have a lighter upper tail except for a handful of extremely large
    sample mean values. So it’s not a very good fit, possibly because of the sample size not
    being large enough.

    You might also like