a)

The estimated regression equation is:

ŷ = 4.94 + 10.49 FundDE + 21.68 FundIE

There is a significant relationship between the dependent variable (5 year average return %)
and the independent variables FundDE and FundIE

State the test hypotheses:

Null hypothesis:
H₀: β₁ = β₂ = 0
That is, there is not a significant relationship between dependent variable (5 year average
return %) and independent variables FundDE and FundIE.
Alternative hypothesis:
Hₐ: Atleast one of the coefficients is not equal to zero
That is, there is a significant relationship between dependent variable (5 year average return
%) and independent variables FundDE and FundIE
According to the output,
 the F-statistic with a numerator degree of freedom of 1
 a denominator degree of freedom of 42,
 corresponding to regression, is 33.45
 the p-value for the F-statistic corresponding to regression is 0.000.
Level of significance:
The given level of significance is α = 0.05.
Rejection rule:
If the p-value ≤ α, then reject the null hypothesis.
Conclusion:
Here, the p-value is less than the level of significance.
That is, p-value (= 0.000) < α (= 0.05)
Thus, the decision is "reject the null hypothesis."

b)

R-squared (R²)

In the given output, R-squared = 61.44%.

Consequently, the independent factors in the regression model help to explain 58.68% of the
variance in the dependent variable. Stated differently, 58.68% of the variability in the dependent
variable can be explained by the model.

evaluation of Model Fit:

The model offers a strong match, as indicated by the R-squared value of 58.68%. Despite its
shortcomings, it accounts for a sizable amount of the variance in the dependent variable. To perform
a thorough evaluation of the model's quality, it's crucial to take into account additional elements
such as the analysis's context and the model's assumptions.

c)
The estimated regression equation that could be used to predict the 5-year average return
given the type of fund, the net asset value and the expense ratio is:

5 year average return % = 1.19 + 6.90 FundDE + 17.68 FundIE – 0.0283 Net Asset
Value + 6.46 Expense Ratio

State the test hypotheses:

Null hypothesis:

H₀: β₁ = β₂ = β₃ = β₄ = 0
That is, there is not a significant relationship between dependent variable (5 year average
return %) and independent variables FundDE , FundIE , the net asset value and the expense
ratio .

Alternative hypothesis:

Hₐ: Atleast one of the coefficients is not equal to zero

That is, there is a significant relationship between dependent variable (y) and independent
variables FundDE (x₁), FundIE (x₂), the net asset value (x₃) and the expense ratio (x₄).

According to the output,

 the F-statistic with a numerator df of 1

 denominator df of 40, corresponding to regression, is 16.25
 the p-value for the F-statistic corresponding to regression is 0.000.

Level of significance:

The given level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is less than the level of significance.

That is, p-value (= 0.000) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

State the test hypotheses:

Null hypothesis:

H₀: β₂ = 0

That is, the addition of the FundIE is not significant.

Alternative hypothesis:

Hₐ: β₂ ≠ 0

That is, the addition of the FundIE is significant.

From the output, it is found that the t-statistic corresponding to FundIE is 5.3315 with
df of 40 and the p-value is 0.000.
Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is less than the level of significance.

That is, p-value (= 0.000) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

Therefore, the data provide sufficient evidence to conclude that the addition of the
independent variable FundIE

is significant.

Therefore, the data provide sufficient evidence to conclude that the addition of the dummy variable
FundIE is significant.

State the test hypotheses:

Null hypothesis:

H₀: β₃ = 0

That is, the addition of the Net Asset Value is not significant.

Alternative hypothesis:

Hₐ: β₃ ≠ 0

That is, the addition of the Net Asset Value is significant.

From the output, it is found that the t-statistic corresponding to Net Asset Value is 0.695 with df of
40 and the p-value is 0.695.

Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is greater than the level of significance.

That is, p-value (= 0.6950) > α (= 0.05)

Thus, the decision is "fail to reject the null hypothesis."

Therefore, the data do not provide sufficient evidence to conclude that the addition of the Net Asset
Value is significant.

State the test hypotheses:

Null hypothesis:

H₀: β₄ = 0

That is, the addition of the Expense Ratio is not significant.

Alternative hypothesis:

Hₐ: β₄ ≠ 0

That is, the addition of the Expense Ratio is significant.

From the output, it is found that the t-statistic corresponding to Expense Ratio is 2.3399 with df of 40
and the p-value is 0.024.

Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is less than the level of significance.

That is, p-value (= 0.024) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

Therefore, the data provide sufficient evidence to conclude that the addition of the Expense Ratio is
significant.

Thus, Net Asset value is not significant and it should be deleted from the estimated regression
equation.

D)

The estimated regression equation;

5 year average return = -4.61 + 8.17 FundDE + 19.52 FundIE + 5.52 Expense Ratio + 5.92 3StarRank
+8.244 4StarRank +6.62 5StarRank

None of the independent variables are deleted from the regression equation.
State the test hypotheses:

Null hypothesis:

H₀: β₁ = 0

That is, the addition of the independent variable FundDE is not significant.

Alternative hypothesis:

Hₐ: β₁ ≠ 0

That is, the addition of the independent variable FundDE is significant.

From the output, it is found that the t statistic corresponding to FundDE is 3.59 with df of 38 and the
p-value is 0.001.

Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is less than the level of significance.

That is, p-value (= 0.001) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

The data provide sufficient evidence to conclude that the addition of the independent variable
FundDE is significant.

State the test hypotheses:

Null hypothesis:

H₀: β₂ = 0

That is, the addition of the FundIE is not significant.

Alternative hypothesis:

Hₐ: β₂ ≠ 0

That is, the addition of the FundIE is significant.

From the output, it is found that the t-statistic corresponding to FundIE is 7.02 with df of 38 and the
p-value is 0.000

Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is less than the level of significance.

That is, p-value (= 0.000) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

Therefore, the data provide sufficient evidence to conclude that the addition of the dummy variable
FundIE is significant.

State the test hypotheses:

Null hypothesis:

H₀: β₃ = 0

That is, the addition of the Expense Ratio is not significant.

Alternative hypothesis:

Hₐ: β₃ ≠ 0

That is, the addition of the Expense Ratio is significant.

From the output, it is found that the t-statistic corresponding to Expense Ratio is 2.13 with df of 38
and the p-value is 0.039.

Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is less than the level of significance.

That is, p-value (= 0.039) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

Therefore, the data provide sufficient evidence to conclude that the addition of the Expense Ratio is
significant.

State the test hypotheses:

Null hypothesis:

H₀: β₄ = 0

That is, the addition of the 3StarRank is not significant.

Alternative hypothesis:

Hₐ: β₄ ≠ 0

That is, the addition of the 3StarRank is significant.

From the output, it is found that the t-statistic corresponding to 3StarRank is 2.10 with df of 38 and
the p-value is 0.043.

Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:
Here, the p-value is less than the level of significance.

That is, p-value (= 0.043) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

Therefore, the data provide sufficient evidence to conclude that the addition of the 3StarRank is
significant.

State the test hypotheses:

Null hypothesis:

H₀: β₅ = 0

That is, the addition of the 4StarRank is not significant.

Alternative hypothesis:

Hₐ: β₅ ≠ 0

That is, the addition of the 4StarRank is significant.

From the output, it is found that the t-statistic corresponding to 4StarRank is 2.11 with df of 38 and
the p-value is 0.042.

Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is less than the level of significance.

That is, p-value (= 0.006) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

Therefore, the data provide sufficient evidence to conclude that the addition of the 4StarRank is
significant.

State the test hypotheses:

Null hypothesis:

H₀: β₆ = 0

That is, the addition of the 5StarRank is not significant.

Alternative hypothesis:

Hₐ: β₆ ≠ 0

That is, the addition of the 5StarRank is significant.

From the output, it is found that the t-statistic corresponding to 5StarRank is 2.11 with df of 38 and
the p-value is 0.042.

Level of significance:

Assume the level of significance is α = 0.05.

Rejection rule:

If the p-value ≤ α, then reject the null hypothesis.

Conclusion:

Here, the p-value is less than the level of significance.

That is, p-value (= 0.042) < α (= 0.05)

Thus, the decision is "reject the null hypothesis."

Therefore, the data provide sufficient evidence to conclude that the addition of the 5StarRank is
significant.

Thus, all the independent variables are significant and no independent variables are deleted from
the regression equation.

E)

The predicted the 5-year average return for a domestic equity fund with an expense ratio of
1.05% and a 3-star Morningstar Rank is 56.057%.

5 year average return = -4.61 + 8.17 FundDE + 19.52 FundIE + 5.52 Expense
Ratio + 5.92 3StarRank +8.244 4StarRank +6.62 5StarRank

The 5-year average return for a domestic equity fund with an expense ratio of 1.05% and a 3-
star Morningstar Rank implies that,

FundDE = 1, FundIE = 0, Expense Ratio = 1.05, 3StarRank = 1, 4StarRank = 0

and 5StarRank = 0.

Thus, the predicted the 5-year average return for a domestic equity fund with an expense ratio
of 1.05% and a 3-star Morningstar Rank is,

5 year average return = -4.61 + 8.17 (1) + 19.52 (0) + 5.52 (1) + 5.92 (1)
+8.244(0) +6.62 (0)

=15.28
PART B

a)

The data related to the wet, noise, buy again and purchase variables for 68 season tires. The
independent variable x1 defines the variable labelled wet, x2 defines the variables labelled
noise and the dependent variable y denoting purchase which is defined as,

y = {
1 if the value of the Buy-Again variable is 7 or greater
0 if the value of the Buy-Again variable is less than 7

Logistic regression equation:

E(y) = 1/(1+e^-(b0+b1x1+...+bpxp))

If the two values of the dependent variable y are coded as 0 or 1, then the logistic regression
equation can be defined as,

E(y) = P(y = 1|x_1, x_2, ..., x_p)

Thus the logistic regression equation relating x1 = Wet performance rating, x2 = Noise
performance rating to y = Purchase is E(y)=1/(1+e−(b0+b1x1+b2x2)).

b)

The estimated logit is g^(x)=−29.82 + 2.69Wet+ 1.210Noise.

Calculation:

Odds:

odds = (P(y=1|x_1, x_2, ..., x_p))/(P(y=0|x_1, x_2, ..., x_p))

Logit:

g(x_1, x_2, ..., x_p) = ln(odds) = b_0 + b_1x_1 + b_2x_2 + ... + b_px_p

The estimated logit is defined as,

g(x_1, x_2, ..., x_p) = b_0 + b_1x_1 + b_2x_2 + ... + b_px_p

c)

Calculation:

If tires have a Wet performance rating of 8 then x1=8 and if tires have a Noise performance
rating of 8 then x2=8.

The estimated logit is,

g(x) = -29.82 + 2.69(8) + 1.21(8)

= 1.38

Now, the estimated value of y is,

ŷ = e^(g(x)) / (1 + e^(g(x)))
= e^(1.38) / (1 + e^(1.38))
≈ 0.798991
Thus, the estimated probability that a customer will probably or definitely purchase a
particular tire again with a wet performance rating of 8 and a Noise performance rating of 8 is
0.798991

d)

If tires have a Wet performance rating of 7 then x1=7 and if tires have a Noise performance
rating of 7 then x2=7.

The estimated logit is,

g(x) = -29.82 + 2.69(7) + 1.21(7)

= -2.52

Now, the estimated value of y is,

ŷ = e^(g(x)) / (1 + e^(g(x)))
= e^(-2.52) / (1 + e^(-2.52))

≈ 0.07446

Thus, the estimated probability that a customer will probably or definitely purchase a
particular tire again with a wet performance rating of 7 and a Noise performance rating of 7 is
0.07446.

e)

Given that wet and noise performance appear to have a significant impact on customer
purchase decisions, as CEO, I will think about giving priority to the development and
promotion of tires with higher performance ratings.