In: Statistics and Probability
Twenty-one mature flowers of a particular species were dissected, and the number of stamens and carpels present in each flower were counted.
x, Stamens | 52 | 68 | 70 | 38 | 61 | 51 | 56 | 65 | 43 | 37 | 36 | 74 | 38 | 35 | 45 | 72 | 59 | 60 | 73 | 76 | 68 |
y, Carpels | 21 | 32 | 29 | 19 | 20 | 30 | 31 | 29 | 20 | 24 | 21 | 30 | 27 | 24 | 28 | 22 | 34 | 26 | 32 | 34 | 35 |
(a) Is there sufficient evidence to claim a linear relationship
between these two variables at α = .05?
(i) Find r. (Give your answer correct to three decimal
places.)
(iii) State the appropriate conclusion.
Reject the null hypothesis, there is not significant evidence to claim a linear relationship.Reject the null hypothesis, there is significant evidence to claim a linear relationship. Fail to reject the null hypothesis, there is significant evidence to claim a linear relationship.Fail to reject the null hypothesis, there is not significant evidence to claim a linear relationship.
(b) What is the relationship between the number of stamens and the
number of carpels in this variety of flower?. (Give your answers
correct to two decimal places.)
= | + x |
(c) Is the slope of the regression line significant at α =
.05?(i) Find t. (Give your answer correct to two decimal
places.)
(ii) Find the P-value. (Give your answer bounds
exactly.)
< p <
(iii) State the appropriate conclusion.
Reject the null hypothesis, there is not evidence of a significant slope.Reject the null hypothesis, there is evidence of a significant slope. Fail to reject the null hypothesis, there is evidence of a significant slope.Fail to reject the null hypothesis, there is not evidence of a significant slope.
(d) Find the 98% prediction interval for the number of carpels that
one would expect to find in a mature flower of this variety if the
number of stamens were 61. (Give your answers correct to one
decimal place.)
Lower Limit | |
Upper Limit |
a)
x | y | (x-x̅)² | (y-ȳ)² | (x-x̅)(y-ȳ) |
52 | 21 | 16.3832 | 36.5737 | 24.4785 |
68 | 32 | 142.8594 | 24.5261 | 59.1927 |
70 | 29 | 194.6689 | 3.8118 | 27.2404 |
38 | 19 | 325.7166 | 64.7642 | 145.2404 |
61 | 20 | 24.5261 | 49.6689 | -34.9025 |
51 | 30 | 25.4785 | 8.7166 | -14.9025 |
56 | 31 | 0.0023 | 15.6213 | -0.1882 |
65 | 29 | 80.1451 | 3.8118 | 17.4785 |
43 | 20 | 170.2404 | 49.6689 | 91.9546 |
37 | 24 | 362.8118 | 9.2880 | 58.0499 |
36 | 21 | 401.91 | 36.57 | 121.24 |
74 | 30 | 322.29 | 8.72 | 53.00 |
38 | 27 | 325.72 | 0.00 | 0.86 |
35 | 24 | 443.00 | 9.29 | 64.15 |
45 | 28 | 122.05 | 0.91 | -10.52 |
72 | 22 | 254.48 | 25.48 | -80.52 |
59 | 34 | 8.72 | 48.34 | 20.53 |
60 | 26 | 15.62 | 1.10 | -4.14 |
73 | 32 | 287.38 | 24.53 | 83.95 |
76 | 34 | 398.10 | 48.34 | 138.72 |
68 | 35 | 142.86 | 63.24 | 95.05 |
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 1177.00 | 568.00 | 4064.95 | 532.95 | 855.95 |
mean | 56.05 | 27.05 | SSxx | SSyy | SSxy |
i) correlation coefficient , r = Sxy/√(Sx.Sy) = 0.5815
ii) Ho: ρ = 0
Ha: ρ ╪ 0
n= 21
alpha,α = 0.05
correlation , r= 0.5815
t-test statistic = r*√(n-2)/√(1-r²) =
3.116
DF=n-2 = 19
p-value = 0.0057
Decison: p value < α , So, Reject
Ho
critical t-value = 2.0930
iii) Reject the null hypothesis, there is significant evidence to claim a linear relationship
b)
sample size , n = 21
here, x̅ = Σx / n= 56.048 ,
ȳ = Σy/n = 27.048
SSxx = Σ(x-x̅)² = 4064.9524
SSxy= Σ(x-x̅)(y-ȳ) = 856.0
estimated slope , ß1 = SSxy/SSxx = 856.0
/ 4064.952 = 0.21057
intercept, ß0 = y̅-ß1* x̄ =
15.24574
so, regression line is Ŷ = 15.25
+ 0.21 *x
c)
i)
Ho: ß1= 0
H1: ß1╪ 0
n= 21
alpha = 0.05
estimated std error of slope =Se(ß1) = Se/√Sxx =
4.309 /√ 4064.95 =
0.0676
t stat = estimated slope/std error =ß1 /Se(ß1) =
0.2106 / 0.0676
= 3.12
ii) 0.001< p value <0.01
iii) Reject the null hypothesis, there is evidence of a significant slope.
d)
X Value= 61
Confidence Level= 98%
Sample Size , n= 21
Degrees of Freedom,df=n-2 = 19
critical t Value=tα/2 = 2.539 [excel
function: =t.inv.2t(α/2,df) ]
X̅ = 56.05
Σ(x-x̅)² =Sxx 4064.95
Standard Error of the Estimate,Se= 4.3086
Predicted Y at X= 61 is
Ŷ = 15.2457 +
0.2106 *61= 28.090
For Individual Response Y
standard error, S(ŷ)=Se*√(1+1/n+(X-X̅)²/Sxx) =
4.423
margin of error,E=t*std error=t*S(ŷ)=
2.539 * 4.423 =
11.2313
Prediction Interval Lower Limit=Ŷ -E =
28.090 - 11.231 =
16.9
Prediction Interval Upper Limit=Ŷ +E =
28.090 + 11.231 =
39.3