代写MBA 2023-2024 Analytics Re-sit Individual Assignment Core Course 1 Credit代写Java编程

MBA 2023-2024

Analytics

Re-sit Individual Assignment

Core Course

1 Credit

Part 1 – Dill Computers

Dill Computers is developing and testing a new computer workstation, Spice101, which it will introduce to the  market  in the  next  6  months.  Spice101  will  be  sold  under  a  three-year warranty covering parts and labour. The company has decided to subcontract the service support for the warranty and has entered negotiations about the support contract with Fixit. Fixit has proposed two different pricing schemes for the subcontract. The first involves the payment of a fixed fee of £1,000,000 and the second a variable fee of £250 per workstation sold, subject to a minimum fee of £500,000. Under both schemes, the payment will be made one year after the introduction of the workstation to the market at which point the product will be replaced by newer models not covered by the warranty service subcontract. At the moment, there  is  uncertainty  about the sales  potential of the  new  workstation. Sales of Spice101  are  expected  to  come  from  two  sources:  (i)  the  successful  closure  by  senior management of a major purchase of 1,000 units by a long standing customer, (ii) the efforts of regional sales offices. Given the state of the negotiations with the long-standing customer, the current estimate of the probability of a successful closure of the major purchase is 60%. Regional sales of Spice101 would be boosted by the successful closure, and the management of  Dill  Computers  has  estimated  the  regional  sales  potential  (in  addition  to  the  major purchase), as shown in the table below.

If major purchase Sales                         Probability	If no major purchase Sales                         Probability
2,000                             25%	500                                20%
3,000                             25%	1,500                             30%
4,000                             25%	2,500                             30%
5,000                             25%	3,500                             20%

(a) Use a decision tree to establish which of the two pricing schemes minimises the expected value of costs for Dill? (30 marks)

(b)  By  how  much  would  the variable fee  have to  change  before your  answer to  part  (a) changes? (10 marks)

(c) By how much would the minimum fee have to change before your answer to part (a) changes? (10 marks)

Part 2 – Shine Bright Like a Diamond

A regression model is required to support the pricing of individual diamonds. The  file ShineBright.xlsx contains the  individual selling  price and various characteristics of  10,000 diamonds. Here is a description of the variables in the dataset:

Price                                     Price in US dollars

Carat                                   Weight in carats

Cut                                      1 (worst) to 5 (best), representing AGS grades fair, good, very good, premium and ideal

Colour                                 1 (worst) to 7 (best), representing GIA grades J to D

Clarity                                 1 (worst) to 8 (best), representing GIA grades I1, SI2, SI1, VS2, VS1, VVS2, VVS1 and IF

Table                                  Width of the top of the diamond as a percentage of its widest diameter

Length                                 Length in mm

Width                                  Width in mm

Depth                                   Depth in mm

DepthPercentage             Depth divided by average of Length and Width

(a) Split the data for the 10,000 diamonds into a training dataset and a testing dataset. Discuss the motivation for splitting the data in this way. (5 marks)

(b) Using only the training dataset, estimate a regression model. Make sure to describe the steps of your modelling approach, and to interpret your model. (35 marks)

(c) Evaluate the accuracy of your model from part (b) in terms of its ability to predict the price of the diamonds in the testing dataset. Compare this with the accuracy of at least one other regression model estimated using the training dataset. (10 marks)

MARKING CRITERIA: QUALITATIVE

	Mark band for core courses	Mark band for all other courses	MARKING CRITERIA
D IST INCT ION	70 and 75	80-100	Superb work. Directly answers the question set. Draws on an extensive range of material. The   answer   demonstrates   the   ability   to   critique   the underlying  concepts,  and/or  shows  total  grasp  of  the relevant frameworks.
		75-79	Excellent work. Directly answers the question set. Draws on a very wide range of material. The answer demonstrates the ability to critique the underlying concepts, and/or shows an excellent grasp of the relevant frameworks.
		70-74	Fine work. Directly answers the question set. Draws on a wide range of material. The answer demonstrates the ability to critique the  underlying  concepts, and/or shows a fine grasp of the relevant frameworks.
PASS	65	65-69	Strong pass. Addresses the question set. Draws on a range of material. The answer demonstrates a strong grasp of concepts and frameworks.
	55 and 60	55-64	Good pass. Addresses the question set. Draws on the main material. The answer demonstrates good but not complete understanding of concepts and frameworks.
	50	50-54	Pass. Addresses the question set. Draws on a limited range of    material.    The    answer    demonstrates    satisfactory knowledge of some of the key concepts and frameworks.
FA IL	45	45-49	Marginal fail. Attempts to answer the question set. Fails to draw on the key material. The answer fails to demonstrate satisfactory knowledge of the key concepts and frameworks.
		0-44	Outright fail. Fails to address the question set. Fails to draw on  the   key  material.  The  answer  fails  to  demonstrate satisfactory    knowledge    of    the    key    concepts    and frameworks.