Business Analytics

This document provides information on business analytics, including the presentation of a report, the adoption of innovative strategies in the retail industry, regression analysis for cost estimation, and classification methods. It also discusses the use of neural networks and clustering in data analytics.

Running head: BUSINESS ANALYTICS 1
Business Analytics
Student’s Name:
Institutional Affiliation

BUSINESS ANALYTICS 2
Business Analytics
Q1
i The presentation of the report is good. The report explains the topic at hand using
writings supported by attractive visualizations such bar graphs and pictures. The
visualizations are of desirable quality and hence can attract readers to read to establish the
information presented in the report. The information provided is also presented in a
professional language builds the trust of the reader on the correctness of the insights.
ii The report shows that majority of retail industry players have adopted innovative
strategies. Therefore, this aids in decision making by motivating firms in the industry to
innovate to avoid losing customers.
The retailers expect a return on their investment in innovation in a short period of time
with around 80% of investors expecting returns within a period of one year. With these
retailers receiving an average return of two dollars for every dollar invested in
innovation, managers need to consider the importance of innovation in raising
productivity and competitiveness.
iii The report discusses the state of the Australian retail industry. The report highlights that
the industry is competitive and participants are keen to enhance efficiencies through the
use of innovation and technology. However, some participants tend to have been left
behind in leveraging technology. The report indicates that the industry is operating above
the country’s average in adopting technology with around 87% of retailers being either
innovative active or improvers while around 3 percent have abandoned technology
completely. The major driver of innovation in the retail sector is the improvement of
efficiency and productivity.

BUSINESS ANALYTICS 3
iv Although the report is good, the number of charts in the presentation should be reduced
without losing information. It is possible to incorporate a few visualizations that are
easily comprehensible and rich in information (Ward, Grinstein & Keim, 2015). This will
make the report shorter and information rich.
Q2
i Regression analysis can be used in cost estimation while predicting the relationship
between costs incurred and the level of business level of activity (Kaplan, 2015). For
instance, a car repair company can use regression analysis to estimate the total cost of
repairing a given number of cars. In this scenario, total variable cost is the dependent
variable while the number of cars repaired becomes the independent variable. Regression
analysis leads to development of an equation in the form of Y=a+bx such that the
dependent variable Y can be estimated given the value of the independent variable X.
ii Height and Weight Dataset
Table 1
Height (cm) Weight (Kg)
1 182 77
2 161 58
3 161 53
4 177 68
5 157 59
6 170 76
7 167 76
8 186 69

BUSINESS ANALYTICS 4
9 178 71
10 171 65
iii. Scatter Graph
Figure 1
155 160 165 170 175 180 185 190
0
10
20
30
40
50
60
70
80
90
R² = 0.418809411615452
Scatter Graph of Weight and Height
Height
Weight
The scatter graph illustrates the relationship between variables (Bell, Bryman & Harley, 2018).
The above scatter graph shows that weight and height have a linear relationship such that taller
the student, the heavier she is. Therefore, tall students on average weigh more than shorter
students..
iv. Compute equation
Y-intercept =60
(178,71) (157,59)

BUSINESS ANALYTICS 5
Change in Y/ change in X = (59-71)/ (157-178) =0.5714
Hence trend line equation is Y=60+0.5714x
v. Calculate R-squared
Height (cm)-
X
Weight (Kg)-
Y
Y_hat
Y_hat-Y
(Y_hat-
Y)^2
Y_bar-
Y
(Y_bar-
Y)^2
1 182 77 163.99 86.99 7567.26 -9.8 96.04
2 161 58 153.71 95.71 9160.404 9.2 84.64
3 161 53 152.00 99 9801 14.2 201.64
4 177 68 149.71 81.71 6676.524 -0.8 0.64
5 157 59 149.71 90.71 8228.304 8.2 67.24
6 170 76 157.14 81.14 6583.7 -8.8 77.44
7 167 76 171.42 95.42 9104.976 -8.8 77.44
8 186 69 166.28 97.28 9463.398 -1.8 3.24
9 178 71 161.71 90.71 8228.304 -3.8 14.44
10 171 65 157.71 92.71 8595.144 2.2 4.84
Mean of Y Y_bar= 67.2 Sum=
83409.02
Sum=
627.6

BUSINESS ANALYTICS 6
R-squared = square root (83409.02/ 627.6) = 11.528%. The model explains 11.528% of variation
in weight. Therefore, it is a poor model for prediction purposes.
vi. Use analytical tool
Output 1
From the Output 1 above, we can conclude that the equation of the regression line is;
Y= -28.2277+0.5581x where;
Y represents weight

BUSINESS ANALYTICS 7
X represents height.
R – Squared
From the output above, it can be depicted that the R-squared value of goodness of fit is
41.8809%. This means that the model explains the variation in weight by 41.8809%. Since
58.1191% of the variation remains unexplained by the model, we conclude that the model does
not adequately explain the variation in the dependent variable. Compared to the calculated
equation and R-squared, the results obtained using the analytical model are better.
QN 3
i Classification predicts categorical class of variables by applying labels of training data to
classify new ones whereas prediction models continuous variables (Aggarwal, 2014). A
researcher will use classification to determine whether a student will pass or Not and use
prediction to estimate the cost of repairing one thousand cars in a garage.
ii Examples of classification methods are discriminant analysis, random forests, logistic
regression, k-nearest neighbors, neural networks, and the naïve Bayes method. These
methods have their unique characteristics (Sajana, Rani & Narayana, 2016). The nature of
variables in the dataset determines the most appropriate method to be applied.
iii

BUSINESS ANALYTICS 8
The algebraic equation for y1 in terms of input values i1, i2, and weights w is.
Y1= bi+w1⋅i1+w2⋅i2+...+wn⋅xn
i Use of Neural Networks
Analysts consider neural networks as classification layer that lies on top of the data being
managed or stored. To classify, unlabeled data is grouped as per their input similarities. Neural
networks can also help in classifying trained labeled data.
ii Clustering
Clustering in data mining involves grouping together related data. The algorithm allocates data
points to various groups some of which are similar and other not (Chicco, Napoli & Piglione, 2018).
In data analytics, clustering can be used to manage data more efficiently. Clustered data are
easily accessible to the right users.
Businesses can also use clustering to segment their customers according to their characteristics.
Market segmentation which is achieved through clustering customers makes sales campaigns
more successful in generating sales (Ernst & Dolnicar, 2018)

BUSINESS ANALYTICS 9
Insurers apply clustering in detecting fraud, and identification of possible risk factors. Insurers
face the risk of moral hazard (Tan, 2018). Therefore, clustering enables them to identify the most
risky clients in order for to charge appropriate rates of premium.
References

BUSINESS ANALYTICS 10
Aggarwal, C. C. (Ed.). (2014). Data classification: algorithms and applications. CRC press.
Bell, E., Bryman, A., & Harley, B. (2018). Business research methods. Oxford university press
Chicco, G., Napoli, R., & Piglione, F. (2018). Comparisons among clustering techniques for electricity
customer classification. IEEE Transactions on Power Systems, 21(2), 933-940.
Ernst, D., & Dolnicar, S. (2018). How to avoid random market segmentation solutions. Journal of Travel
Research, 57(1), 69-82.
Kaplan, R. S., & Atkinson, A. A. (2015). Advanced management accounting. PHI Learning.
Sajana, T., Rani, C. S., & Narayana, K. V. (2016). A survey on clustering techniques for big data
mining. Indian journal of Science and Technology, 9(3), 1-12.
Tan, P. N. (2018). Introduction to data mining. Pearson Education India.
Ward, M. O., Grinstein, G., & Keim, D. (2015). Interactive data visualization: foundations, techniques, and
applications. AK Peters/CRC Press.