Assignment Brief
BEM2039 BUSINESS ANALYTICS IN PRACTICE T1 – 2022/23
Summative Assessments:
The Summative Assessments are assessed by individual coursework exercise worth 60% of your final mark and a group presentation (video), worth 40% of your final mark, as follows:
1) Individual coursework exercise
Length : 2,000 words (± 10%)
Weight : 60% of the overall module mark
Deadline: 18.12.23 (Monday) at 15:00 via eBART
2) Group Presentation
Length : 15 minutes (± 10%)
Weight : 40% of the overall module mark
Deadline : 18.12.23 (Monday) at 15:00 via eBART
Summative Assessments 1:
Please answer all 5 questions. Neatly type your answers on a Word document and convert it to PDF before submitting. This is to ensure that all diagrams drawn, equations or outputs of analysis from the R software remain intact. Ensure that you include detailed explanations, examples and workings on how you derive the answers.
1) Explain the importance of analytics and data science in business.
2) Explain the concept of a random variable and make at least two examples involving a discrete random variable.
3) You have recently been appointed as a management consultant with the purpose of providing solutions to a project involving the establishment of a new Music Store in a UK city. In particular, you have been tasked to evaluate a set of variables that, in your expert opinion, can affect the sales of this store.
a. Construct a model of the type y=f(x), being y the sales revenue and x the matrix of the independent variable (at least 3). Explain the direction of the expected correlation of each of the variables included in your model;
b. Suppose that two sets of monthly data (related to the period January -- December 2020) are available for a music store considered similar to the one you have in mind. They are sales revenues in thousands of dollars (the variable y) and number of people, between the age 15 and 20, who visited the store in each month (the variable x). Their numeric values are as follows:
y=(16230,21456,30344,18056,11759,19458,18478,15112,9836, 31000,19450,20016);
x=(5001,9233,12254,7643,3365,7297,9487,5010,4416,16054,
9001,11017).
Using the software R, plot both the variables and compute their mean, standard deviation and variance. Tell whether x and y are correlated or not and explain why.
c. Using the software R, build a regression model (use the command “lm()”) involving the above defined variables y and x. Explain the difference between dependent and independent variables and their role in a regression model. Also comment on the following statistics: R2,R2adj, standard errors of the coefficient(s). Finally,
d. Using the software R, build an autoregressive model (use the command ar()) of order 1 for the variable y (sales revenues) and then, according to it, predict its future values related to the months of January and February 2021. Explain the concept of autoregressive models and write the related equation.
4) Given the following two vectors, z=(16,22,30) and w=(12,11,20), compute by hand the covariance and the correlation coefficient
5) Data ethics, governance and professional responsibility are key topics which can be regarded as a set of rules any business analyst should follow. Explain why it is important to strictly observe such rules and why it is dangerous to breach them.
Individual coursework (60% of the credits): marking grid
|
70% + |
60 – 69% |
50 – 59% |
40 – 49% |
Below 40% |
|
Understanding of problems and context – 15 % |
||||
|
Demonstrates ability to deal with and summarize a range of complex issues, formalizing effectively business problems and possible strategies |
The topics and questions being explored are clearly identified. Context is clear and is properly discussed |
Demonstrates ability to analyse complex issues and make appropriate judgements |
Demonstrates little ability to analyse complex issues. Significant inconsistencies in summarizing and formalizing business analytics problems |
Very little or no demonstrable ability in formalizing/summarizing business analytics problems |
|
Personal attitude towards ethical issues in Business Analytics - 25 % |
||||
|
Clear evidence of deep understanding of morality principles in business analytics |
Ethical principles in business analytics are properly discussed |
Sufficient knowledge of the moral principles driving the analysis of business data |
There are noticeable omissions in ethical considerations in conducting business data analysis |
Ethical issues have not been considered or adequately discussed |
|
Knowledge of analytical tools - 25% |
||||
|
Outstanding level of comprehension of statistical modelling and main statistical functions |
Clear understanding of the analytical tools and the related concepts. |
Sufficient knowledge of statistical functions and their applications. Demonstrate ability in statistical modelling |
Poor ability in handle statistical functions and inconsistency in model building procedures |
Very little or no demonstrable ability in model building procedures. Statistical functions are inappropriately used. |
|
Data analysis skills - 35% |
||||
|
Excellent application of the analytical tools and outstanding capabilities in transforming data into knowledge |
The use of data analysis is clearly understood as well as the principles to derive useful information from the data |
Demonstrates ability in conducting data analyses but lack of perspective in terms of usability of the potentially usable information |
Lack of ability in generating a valuable flow of information from the data. The analysis conducted has not been correctly performed or is not properly justified |
Poor understanding of the relevant steps of a data analysis. |
Summative Assessments 2 - Group presentation: 15 minutes video
Groups of 4/5 people are formed according to the interest showed by the students towards the following three topics:
1) BUSINESS ANALYTICS IN PRACTICE AND STOCK MARKET USING R
2) BUSINESS ANALYTICS IN PRACTICE AND BIG DATA USING R
3) BUSINESS ANALYTICS IN PRACTICE AND SALES FORECASTING USING R.
Each student is required to equally contribute to all the steps leading to the final presentation which will be delivered by the group leader (selected by the members of each group).
Group 1
BUSINESS ANALYTICS IN PRACTICE AND STOCK MARKET USING R
In the video the following topics should be outlined:
1) The R software and its importance for a business analyst.
2) Explain how to install the software R and the required libraries.
3) Define the stock market and the differences between technical and fundamental analysis.
4) Using the library “quantmod”, select three stocks from different industries, e.g. hi-tech, manufacture, life insurance, commercial banking, and
a. generate the OHLC dataset for each of the chosen stocks and explain the meaning of the acronym OHLC;
b. generate bar-charts for each of the stocks and explain the different dynamics at play;
c. extract the Adjusted Price for each of the stocks and compute the covariance and the correlation coefficient between all the pairs;
d. comment on a possible investment strategy according to the correlation coefficients found.
Group 2
BUSINESS ANALYTICS IN PRACTICE AND BIG DATA USING R
In the video the following topics should be outlined:
1) The R software and its importance for a business analyst.
2) Explain how to install the software R and the required libraries.
3) Define Big Data and explain its impacts on the Business World.
4) Explain what Google Trends is, its advantages and limitations.
5) Select at least 3 keywords consistent with a marketing problem of your choice and download the monthly data of these keywords from the Google Trends website (https://trends.google.com/trends). Import the file in R using the command “read.table”.
6) Use the libraries “sma” and “forecast”
a. generate one single plot where each of the chosen keyword is depicted;
b. generate a suitable smoothed version of each keyword by choosing the smoothing parameter “order” embedded in the command “ma”;
c. compute the covariance and the correlation coefficient for each pair;
d. select the most informative keyword and justify your choice.
Group 3
BUSINESS ANALYTICS IN PRACTICE AND SALES FORECASTING USING R
Choose one out of the three datasets downloadable from my ELE page.
In the video the following topics should be outlined:
1) The R software and its importance for a business analyst.
2) explain how to load the chosen data set in R.
3) print the first 6 data and generate summary statistics through the command “summary()”. Comment on these statistics.
4) explain why the data format is a vector as opposed of a scalar and a matrix.
5) generate the “sales versus time” plot.
6) transform the data into a time-series with frequency=12 (months) and starting date January 1975.
7) decompose the data into three components: trend, seasonal, irregular, using the R command “decompose()”. Plot the results and comment the resulting graph
8) apply the command “ar()” to model the data according to an autoregressive model and generate a 5-step ahead predictions according to 3 different autoregressive parameter settings of your choice (use the R command “forecast”)
9) comment on the obtained predictions.
Video presentation (40% of the credits): marking grid
|
70% + |
60 – 69% |
50 – 59% |
40 – 49% |
Below 40% |
|
Introduction – 20% |
||||
|
The topics and the questions being explored are clearly identified. Context is clear and has been appropriately explored |
The topics are well summarized. Questions are clearly commented and well introduced |
The topics are clearly reported but questions to be explored are not well-framed or show inconsistencies |
The topics are clear, but there are important gaps and omissions |
The topics are not evident, or are poorly described |
|
Data Analysis - 35% |
||||
|
The statistical tools have been applied correctly and accompanied with a strong explanation of their validity |
The analysis has been conducted using the right approach and good explanations of the applied procedures are provided. Good understanding of the limitations of the applied techniques |
The relevant steps have been undertaken with few errors. Some recognition of the limitations of the analytics undertaken |
Lack of explanations of the applied procedures. Some conceptual errors are present |
Analysis is incomplete, unclear or there are significant errors |
|
Conclusion - 30% |
||||
|
Appropriately justified and well evidenced conclusions have been communicated clearly. Good evidence of the limitations of the project
|
Informed and well-presented conclusions. Some considerations of the limitations of the project |
Conclusions are mostly correct but are not fully justified |
Some partial conclusions, but some significant inconsistencies or omissions |
Little or no evidence of critical reflection |
|
Visualization - 15% |
||||
|
Creative and thought-provoking visualizations that are accurate and fairly represent the nature of the data/analysis they describe.
|
Visualization has been carefully considered as an integral part of the video presentation. Slides are well made and impactful. |
Appropriate visualizations that are supportive of the narrative. |
Presentation is basic but appropriate. However, the overall impact on the audience is very limited |
Video presentation is scarcely appealing due to the overall poor visual content |
Please, bear in mind the following:
If there is sufficient evidence that a student is not participating adequately in the group work, the module lead may seek the approval of the Director of Education to remove them from the group work assessment. In this instance the student will be deemed to have failed the first attempt at the assessment and will have the opportunity to complete an individual assessment in the referred/deferred assessment period. This assessment will be capped at the pass mark (50% for PG).