Running head: REGRESSION ANALYSIS 1

Regression Analysis

Name

Institution affiliation

Date
REGRESSION ANALYSIS 2

1. Please explain what a binomial distribution is. What is the symmetry property of
this distribution? How do you find the mean, the variance, and the standard
deviation of this distribution? Provide a simple example of binomial distribution.
A Binomial distribution is one of the most common distributions in statistics. As opposed to

continuous distribution, the binomial distribution represents the probability for x successes in n

trials, given a success probability p for each trial (Dimotikalis, 2015). There are properties or

conditions that distribution must hold for it to be considered binomial. These properties include a

fixed number of observations, independent observations, the outcome of the observation is either

success or failure, and the probability of "success" p is the same for each outcome (Arkes, 2019).

A common example of binomial distributions is the tossing of a coin. The expected outcome is

either “head” or “tail." The probability of each outcome is 0.5.

In binomial distribution,” p” represents the probability of each outcome, whereas “n” represents

the total number of observations.

Mean for this distribution is given as;

Mean=np

Variance=np(1-p)

Standard deviation=√ np(1− p)

2. For this assignment, please retrieve annual rate of return for a major security
within the last 20 years, do the same for a major financial index then do a regression
of the annual rate of return of the security on the annual rate of return of the
financial index to obtain a level of volatility of the security relative to volatility of the
market (the financial index).
The final regression model is; Y=0.0023322+0.4763907X.

The p-value of the model is 0. 0419. this p-value is less than the alpha level, meaning that the

model is significant.
REGRESSION ANALYSIS 3

Coefficient of determination=0.21

Scatter plot

scatter plot
0.6
Annual rate of return of IBM security

0.5
0.4
0.3
f(x) = 0.476390725074616 x + 0.00233217071289788
0.2
R² = 0.210540597903979
0.1
0
-0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8
-0.1
-0.2
-0.3
Annual rate of return of Dow Jones Index

From the look of the scatter plot, the model is not valid.

Normal Probability Plot of Residuals

0.30000

0.20000

0.10000
Residual

0.00000

-0.10000

-0.20000

-0.30000

-0.40000
-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
Normal Score
REGRESSION ANALYSIS 4

The data is not normally distributed. Normal distribution will have its points or scores along the

line.

Revisiting the regression model, the regression model is significant and valid. This means that

the market is outperforming, within the range of volatility.

3. Explain the non-parametric sign test. What is the objective of this test and how the
corresponding hypotheses are formulated? What is the approach to test the
hypotheses using the binomial distribution for both non-directional and directional
cases?
The sign test is a non-parametric test that is used to test the null hypothesis that the median of a

certain group is equal to a certain value. When two groups are involved, the sign test is used to

test the null hypothesis that the differences between median are equal to zero (Sen & Srivastava,

2012). The binomial distribution can be used to compute the p-values for this test. Whenever we

have a two-sided hypothesis test, for example, given with n = 12 and p=0.5. The two-sided p-

value for the test is p-value = 2*P(x < 3) (which is equivalent to p-value = P(x < 3) + P(x > 9)).

4. A survey of 40 home prices in a metropolitan area has the following results.

 Null and Alternative Hypotheses

The following null and alternative hypotheses need to be tested:

Ho: Median of Differences = 0

Ha: Median of Differences ≠ 0

+¿ ¿ −¿¿
The test statistic is X = min{ X ,X }=min{1,13}=1.

Rejection Region

The critical value for the significance level provided, and the type of tail is x ¿ = 2.
REGRESSION ANALYSIS 5

Decision about the null hypothesis

Since, in this case, X=1≤ x ¿=2, there is enough evidence to claim that the population median of

differences is different than 0, at the 0.05 significance level.

+¿¿ −¿¿
The provided number of positive signs is  x = 27, and the number of negative signs is x =

13

Null and Alternative Hypotheses

The following null and alternative hypotheses need to be tested:

Ho: Median of Differences = 0

Ha: Median of Differences ≠ 0

The test statistic is X = min{ x +¿¿, x−¿¿ } =min{27,13}=13.

Observe that the sample size n = x +¿¿ + x−¿¿ = 27 + 13 = 40 >25 is large enough to use normal

approximation, so a z-statistic will be used.

Rejection Region

Based on the information provided, the significance level is α=0.05, and the critical value for a

two-tailed test is zc=1.96.

The rejection region for this two-tailed test is R={z:∣z∣>1.96}

Test Statistics

The z-statistic is computed as follows:

x+ 0.5−n /2 13+0.5−40/2
z= = =−2.055
√ n /2 √ 40 /2

Decision about the null hypothesis

Since it is observed that |z| = 2.055>zc=1.96, it is then concluded that the null hypothesis is

rejected. Using the P-value approach: The p-value is p=0.0398, and since p=0.0398<0.05, it is

concluded that the null hypothesis is rejected. It is concluded that the null hypothesis Ho

is rejected. Therefore, there is enough evidence to claim that the population median of

differences is different than 0, at the α=0.05 significance level.

References

Arkes, J. (2019). Regression Analysis: A Practical Introduction. London, England: Routledge.

Dimotikalis, Y. (2015). Fitting Binomial Distribution to Online Rating Data: TripAdvisor

Ratings in Crete Island: SMTDA 2014 conference presentation. ISAST.

Sen, A., & Srivastava, M. (2012). Regression Analysis: Theory, Methods, and Applications.

Berlin, Germany: Springer Science & Business Media.