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

Welcome to Scribd!

  • 27 views

    Tugasan/Assignment 5 (20 Markah/marks) : Serial No

    Uploaded by

    Nurul Fadzillah
    The document presents a table with data on the number of lines of code (LOC), x, and time taken in minutes, y, for 5 programs by a programmer. It then asks to find the least squares line relating y to x, construct an ANOVA table for the linear regression, and determine whether LOC is useful for predicting time taken. The least squares line is found to be y = 0.665070774x + 19.18089092. The ANOVA table is constructed and the F ratio is greater than the critical value, so LOC is determined to be useful for predicting time taken to finish a program.

    Copyright:

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

    Tugasan/Assignment 5 (20 Markah/marks) : Serial No

    Uploaded by

    Nurul Fadzillah
    27 views4 pages
    The document presents a table with data on the number of lines of code (LOC), x, and time taken in minutes, y, for 5 programs by a programmer. It then asks to find the least squares line relating y to x, construct an ANOVA table for the linear regression, and determine whether LOC is useful for predicting time taken. The least squares line is found to be y = 0.665070774x + 19.18089092. The ANOVA table is constructed and the F ratio is greater than the critical value, so LOC is determined to be useful for predicting time taken to finish a program.

    Original Description:

    stats

    Original Title

    Stats Asg 5

    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 presents a table with data on the number of lines of code (LOC), x, and time taken in minutes, y, for 5 programs by a programmer. It then asks to find the least squares line relating y to x, construct an ANOVA table for the linear regression, and determine whether LOC is useful for predicting time taken. The least squares line is found to be y = 0.665070774x + 19.18089092. The ANOVA table is constructed and the F ratio is greater than the critical value, so LOC is determined to be useful for predicting time taken to finish a program.

    Copyright:

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

    Tugasan/Assignment 5 (20 Markah/marks) : Serial No

    Uploaded by

    Nurul Fadzillah
    The document presents a table with data on the number of lines of code (LOC), x, and time taken in minutes, y, for 5 programs by a programmer. It then asks to find the least squares line relating y to x, construct an ANOVA table for the linear regression, and determine whether LOC is useful for predicting time taken. The least squares line is found to be y = 0.665070774x + 19.18089092. The ANOVA table is constructed and the F ratio is greater than the critical value, so LOC is determined to be useful for predicting time taken to finish a program.

    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 4

    Tugasan/Assignment 5 [20 markah/marks]

    Jadual di bawah menunjukkan bilangan baris kod (LOC), x, dan masa yang diambil
    untuk menyiapkan 5 (lima) aturcara tersebut dalam minit y oleh seorang pengaturcara..
    The table below show the number of lines of codes (LOC), x, and the time taken to
    finish 5 (five) programmes in minutes, y, by a programmer.

    1 2 3 4 5 ∑x ∑ x2
    Bilangan baris 10 20 30 40 49 149 5401
    kod, x
    Line of Codes
    (LOC), x
    Masa dalam 23 35 41 46 50 195 8051
    jam,y
    Time in minutes,
    y
    xy 230 700 1230 1840 2450 6450

    a. Andaikan bilangan baris kod, x dan masa dalam minit, y berhubungan secara
    linear. Cari garis kuasadua terkecil yang menghubungkan y dengan x.
    Assume that the LOC,xand time in minutes, y are linearly related. Find the
    least-squares line relating y to x. [5m]

    Serial No. Line of Codes Time in X2 xy


    (LOC), x minutes , y
    1 10 23 100 230
    2 20 35 400 700
    3 30 41 900 1230
    4 40 46 1600 1840
    5 49 50 2401 2450
    ∑x= ∑y= ∑x2 = ∑xy=
    149 195 5401 6450

    Say, the linear fit be,


    y = ax + b

    where, a and b are unknown constants to be determined from given data.


    Therefore, for least square fitting, we have,

    ∑y = a∑ x + nb
    ∑xy = a∑ x 2 + b∑ x

    1
    Where , n is the number of data(here n= 5)
    So, we have
    195 = 149a + 5b
    6450 = 5401a + 149 b

    Solving for a and b,

    a = 0.665070774, b= 19.18089092

    So, the least square linear fit is:

    y = 0.665070774 x + 19. 18089092

    1
    b. Bina jadual ANOVA bagi regresi linear.
    Construct the ANOVA table for the linear regression. [8m]

    Fitted y (y) Error,e e2 (yi- y) (yi- y)2


    -2.8316 -2.8316 8.017951 -13.1684 173.4068
    2.517694 2.517694 6.338781 -6.51769 42.48033
    1.866986 1.866986 3.485636 0.133014 0.017693
    0.216278 0.216278 0.046776 6.783722 46.01888
    -1.76936 -1.76936 3.130631 12.76936 163.0565
    ∑ yi =¿ ∑e = ∑ e2= ∑ (¿ y i− y )¿= ∑ (¿ y i− y )¿2=
    195 7.4 E-08 21.01978 -7.4E-08 424.9802

    Fitted y is calculated from the obtained least square linear fit, putting corresponding
    value of Line of Codes(LOC),x
    y = 0.665070774 x + 19. 18089092
    Error e is given by,

    e = (y-yi )

    Mean, y is given by, y = ∑ yi /5

    ANOVA table
    df SS MS F test ρ - F crit
    value
    Regressio 1 424.9802 424.9802/1 = 424.9802/ .004405 10.1280
    n 424.9802 7.006593333
    (ESS) =60.65432654
    Residual 3 21.01978 21.01978/3 =
    (RSS) 7.006593333
    Total 4 446
    (TSS)

    1
    c. Tentukan sama ada bilangan baris kod (LOC) adalah berguna bagi meramalkan
    masa untuk menyiapkan aturcara?
    Determine whether the line of codes (LOC) is useful in predicting the time taken
    to finish the program?

    [7m]
    Say, the line of codes (LOC) is not useful in predicting the time taken to finish the
    program. Then, we test this hypothesis for statistical significance. So,

    H0: The time taken to finish the program is not related to the number of line of codes
    (LOC)

    H1: The time taken to finish the program is related to the number of line of codes
    (LOC)

    From ANOVA table, we see that the F ratio for linear regression Ftest is higher than
    the Fcrit value at 5% level of significance with (1,3) df . So, we may reject the null
    hypothesis and conclude that the time taken to finish the program has linear
    relationship to the number of line of codes.

    So, yes, the line of codes (LOC) is useful in predicting the time taken to finish the
    program.

    You might also like