I-MTH203代写、代写Java，Python程序设计

日期：2022-04-15 10:59

Couse work I-MTH203 (S2, AY2021-22) Release Date: April 11 Submission: 5:00pm April 17 via LMO

Note: 1. You are allowed to use computer programming and/or Excel, but codes and/or Excel worksheets and/or snapshots of computations should be provided to gain higher marks. 2. We only accept submissions in the format of pdf, docx, xlsx (i.e.,Excel).

Question 1. [40 marks] A time series dataset is provided in the Excel file “data.xlsx”

which contains 156 data points.

(1) Describe the dataset. Consider various aspects, for example the graph, the

“pattern”, the trend, etc.

(2) Treat the data points as they follow a linear trend model. Generate

157,158, … ,208 by using Double Exponential Smoothing method with S0 = 30, B0 = 10,α = 0.3,β = 0.1 and present the forecasts and real

observations in one graph. Round your answers to 3 decimal places.

Conceptually do you think this is appropriate for this particular dataset? Why?

(3) Use a different way to produce 157,158, … ,208 and make a comparison

with your result in (2), provide sufficient explanation/reasoning. Again present

the forecasts and real observations in one graph. Round your answers to 3

decimal places.

Question 2. [30 marks] Caspar owns a Company which manufactures electronic chips

and his business partner, Martin, would like to order 200,000 chips, which would

make the whole company work full time for a week, each chip profits $5 but if

Caspar couldn’t finish, he has no gain and needs to pay liquidated damage $500,000;

or Martin orders 100,000 chips, each profit $4, in which case the company would

work part-time, the liquidated damage is $150,000. Also Caspar could choose not to

take any orders from Matin and let the company not work for now. Due to the recent

spread of COVID-19 in the city, all the companies are monitored by the government.

So there is a 50% chance that a company with full time operations will be urged to

close temporarily and 20% for companies with part time operations. As an

alternative, companies could conduct a nucleic acid testing (costing $10,000) before

their productions, the ones with “all negative” would be allowed to work (regardless

of full-time or part-time) with probability 0.65, if one or two positive cases were

detected, companies working full-time will be stopped and the ones working

part-time would keep working with probability 0.6. All the companies with more

than two positive cases would be asked to close right away. Suppose that Caspar

wishes to generate the optimal decisions using decision tree method.

(1) Draw the draft tree to present the logic of the problem.

(2) Solve the problem, make additional information if necessary.

Question 3 (DIY) [30 marks]. (1) Create a Markov Decision Process (MDP) question

that contains 3 actions, each with a 3 by 3 probability transition matrix and a 3 by 3

cost matrix. Suppose we make decisions for 4 stages. Introduce the background of

the problem (rather than just giving maths) and ensure that solution of the question

is non-trivial, i.e., it would not always lead to the same action in all stages. (2) Solve

the question you create in (1), show your work.