Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 151 views

    Assignment3 Final

    Uploaded by

    Kunal Nanak Lalchandani
    The document contains analysis of survey data from customers of MyTVLab who used either PAF or Conbanco as a payment medium. It finds that: 1) The proportion of customers using PAF was 0.44 and Conbanco was 0.56, with a total sample of 50. 2) A z-test finds no significant difference between these proportions at the 95% confidence level. 3) Additional questions calculate probabilities for test score ranges using normal approximations. Sample size affects the probability, with a larger sample of 100 giving a higher probability than samples of 90.

    Copyright:

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

    Assignment3 Final

    Uploaded by

    Kunal Nanak Lalchandani
    151 views4 pages
    The document contains analysis of survey data from customers of MyTVLab who used either PAF or Conbanco as a payment medium. It finds that: 1) The proportion of customers using PAF was 0.44 and Conbanco was 0.56, with a total sample of 50. 2) A z-test finds no significant difference between these proportions at the 95% confidence level. 3) Additional questions calculate probabilities for test score ranges using normal approximations. Sample size affects the probability, with a larger sample of 100 giving a higher probability than samples of 90.

    Original Title

    Assignment3_final.docx

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document contains analysis of survey data from customers of MyTVLab who used either PAF or Conbanco as a payment medium. It finds that: 1) The proportion of customers using PAF was 0.44 and Conbanco was 0.56, with a total sample of 50. 2) A z-test finds no significant difference between these proportions at the 95% confidence level. 3) Additional questions calculate probabilities for test score ranges using normal approximations. Sample size affects the probability, with a larger sample of 100 giving a higher probability than samples of 90.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    151 views4 pages

    Assignment3 Final

    Uploaded by

    Kunal Nanak Lalchandani
    The document contains analysis of survey data from customers of MyTVLab who used either PAF or Conbanco as a payment medium. It finds that: 1) The proportion of customers using PAF was 0.44 and Conbanco was 0.56, with a total sample of 50. 2) A z-test finds no significant difference between these proportions at the 95% confidence level. 3) Additional questions calculate probabilities for test score ranges using normal approximations. Sample size affects the probability, with a larger sample of 100 giving a higher probability than samples of 90.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    of 4

    Question 1 (a)

    Customers of MyTVLab using PAF as a payment medium from the sample = 22

    Customers of MyTVLab using Conbanco as a payment medium from the sample = 28

    Total customers in the sample= 50

    Therefore, p̂=22/50 = 0.44 and q̂ =28/50=0.56

    Using Hypothesis testing,

    H0: p=q=0.5

    H1: p≠q

    n=50, np = 0.44*50 = 22 which Is greater than 5

    nq = 0.56*50= 28, which is greater than 5

    Since both of them (np and nq) are greater than 5, normal approximation can be assumed.

    𝑝̂−𝑞̂ 0.44−0.56 −0..06


    Z= 𝑝𝑞 = 0.56 = 0.0707 = 0.8485
    √( ) √(0.44∗ )
    𝑛 50

    Assuming the confidence level as 95%, we can get α= (1-0.95) = 0.05

    Now, from the z table, if our z value lies inside the green region, we will not reject the null
    hypothesis H0.
    The z value at -0.025 and 0.025 comes out to be -1.96 and +1.96 respectively. Now since our
    calculated z value is between these two limits, we will not reject the null hypothesis H0.

    Therefore, the speculation by Mr. Rajan that there are unequal number of users of PAF and
    Conbanco is void which can be said with 95% confidence level.

    Question No: 2 – Exercise no.25, page no. 321

    2.a

    Given,

    Sample size of test takers (n) = 90

    Mean value for critical reading test (µ)= 502

    Standard deviation (sigma) = 100

    Required probability is P[492 ≤ xbar ≤ 512] = P[492 ≤ (xbar - µ)/(sigma/sqrt(n)) ≤ 512]

    On substituting the values, we get values of z as -0.948 and +0.948

    From the values of z, the probability lies between 0.1711 and 0.8289

    Therefore, the final probability is 0.8289 – 0.1711 = 0.6578

    2.b.

    Given,

    Sample size of test takers (n) = 90

    Mean value for Mathematics (µ)= 515

    Standard deviation (sigma) = 100

    Required probability is P[495 ≤ xbar ≤ 515] = P[495 ≤ (xbar - µ)/(sigma/sqrt(n)) ≤ 515]

    On substituting the values, we get values of z as -0.948 and +0.948

    From the values of z the probability lies between 0.1711 and 0.8289

    Therefore, the final probability is 0.8289 – 0.1711 = 0.6578

    The probability of mathematics test upon comparing with critical reasoning test is the same because
    of the same sample size and the same standard deviation.

    2.c.

    Given,

    Sample size of test takers (n) = 100

    Mean value for writing (µ) = 494


    Standard deviation (sigma) = 100

    Required probability is P[484 ≤ xbar ≤ 504] = P[484 ≤ (xbar - µ)/(sigma/sqrt(n)) ≤ 504]

    On substituting the values, we get values of z as -1 and 1

    From the values of z, the probability lies between 0.1587 and 0.8413

    Therefore, the final probability is 0.8413 – 0.1587 = 0.6826

    The probability of the writing test is more when compared to the probabilities of mathematics and
    critical reading. It is because of the large sample size in writing test compared to that of the other
    two.

    Question No: 3 – Exercise no.30, page no. 362

    Q.no. 30

    Given,

    Confidence interval value is 95%

    On the distribution curve the limits for z will be -0.025 and +0.0.25

    The value of z will be -1.96

    Standard deviation (sigma) = 2000

    Error margin (xbar - µ)= 100

    Sample size (n) =? (Need to find out)

    Z = -1.96 = (xbar - µ)/(sigma/sqrt(n))

    -1.96 = -100/(2000/sqrt(n))

    Solving for n we get value of n = 1536.64

    A sample size of 1537 should be used for the survey.

    You might also like