MATH6178 Forecasting

COURSEWORK BRIEF:

Module Code: MATH6178

Assessment: Individual Coursework

Weighting: 100%

Module Title: Forecasting

Module Leader: Steffen Bayer

Submission Due Date: @ 16:00 27th August 2021

Word Count: 2000

Method of Submission: Electronic via Blackboard Turnitin ONLY

(Please ensure that your name does not appear on any part of your work)

Any submitted after 16:00 on the deadline date will be subject to the standard University late penalties (see below), unless an extension has been granted, in writing by the Senior Tutor, in advance of the deadline.

University Working Days Late:	Mark:
1	(final agreed mark) * 0.9
2	(final agreed mark) * 0.8
3	(final agreed mark) * 0.7
4	(final agreed mark) * 0.6
5	(final agreed mark) * 0.5
More than 5	0

This assessment relates to the following module learning outcomes:

LO1 Formulate time series models and construct Python-based versions.

LO2 Use Python functions built in various libraries to fit and analyse such models to data

LO3 Appreciate both the capabilities and the limitations of such computer-based techniques

LO4 Produce well-structure assignment reports describing problem formulation and solution 3

Coursework Brief

1. Background and analysis

You have been tasked to forecast a number of key economic indicators for Singapore until March 2022. The data can be obtained from the Department of Statistics Singapore (SingStat). In addition to forecasts of four key indicators a recommendation is sought as to whether these economic indicators can be used to forecast the applications for bankruptcies.

How to getthe data. Fromthe SingStat tablebuilder at you can download the required time series as detailed below.(https://www.tablebuilder.singstat.gov.sg/publicfacing/mainMenu.action; searching for the Code in the search box will make it easy to find the right dataset)

	Code	Title	Variable
1	M700051	Exchange Rates, (Average For Period), Monthly	US Dollar
2	M601641	Retail Sales Index, (2017 = 100), At Current Prices, Monthly	Total
3	M212751	Singapore Manufactured Products Price Index, By Commodity   Section (1-Digit Level), Base Year 2018 = 100,	Singapore Manufactured Products Price   Index -Overall Items
4	M355271	Index Of Industrial Production (2019 = 100), Monthly	Total
5	M890551	Number Of Bankruptcy Applications, Orders Made And Discharges, Monthly	Applications for Bankruptcy

1.1. Tasks. You have to decide how long a time series you want to use. You will have to use yourown judgmentin inspecting and preparing the databefore carrying out any technical analysis. The analysis is in three parts:

(a) You are asked to take time series 1 to 4 separately and to forecastmonthly behaviouruntil March 2022, using exponential smoothing-type forecastingmethods.

(b ) In, addition you are asked to fit an ARIMA model to one of the time series 1 to 4, for analysisin which you compare the use ofthe ARIMA forecastingmethod and an appropriate exponential smoothing technique. You shouldmake a recommendation as to future use of ARIMA on this time series.

(c) You are finally asked to use time series 1 to 4 as explanatory variablesin amultivariate regression model to forecast bankruptcy applications.

(d) Develop amultiple regressionmodel, use itforthe prediction of bankruptcy applications until march 2022, and report on whether you think the model issatisfactory ornot.

2. What you must produce

You must produce a technical report describing all the analysis done to selectthe mostsuitable forecastingmethod, aswell asthe results obtained. The reportmust be accompanied by the codes used to perform the technical analysis, as well as the resulting graphs. More details on each ofthe aspects ofthe work are given in the nextsubsections.

2.1. The technical report. The technical report must follow the structure described in Subsection 2.4. It should address the three parts of the analysis: exponential smoothing, ARIMA, and regression. For each part, give details ofthe preliminary analysis, datapreparation,models chosen and analysis carried out. Also describe why eachmodel was built and explain the analysis carried out, including an evaluation ofthe effectiveness ofthe models.

2.2. Python codes. You must also prepare and submit the Python codes(as .PY files) that you use to generate the results that will be included in your technical report. If any preliminary operations on your data are needed before applying/developing a Python code for your analysis, it is fine to include this in the corresponding excel file containing your data sets. However, you must complete all the main tasks of your analysis using Python. You can use the codes from the course, use different ones or develop your own. Marking on this aspect of your work will not be based on how well you can program in Python, but rather on the functionality of your codes and their relevance in the corresponding analysis.

To help us easily know what you do in each code, youmust produce a single page document, asAppendix A to yourtechnical report,to give a brief description of what each of your codes does. If you do any preliminary operations on yourdata in the excel file containing yourdata set, a line ortwo should also be included to describe this.

2.3. Analysis and forecast graphs. You are expected to produce graphsto illustrate your analysisin the technical report. Do not include these graphsin the main part of the report(Sections 1- 3; see detailsin nextsubsection), butrather, put all ofthemin AppendixB. You are allowed up to 10 pagesfor the graphs produced for your analysis.Organise the graphsin 3main parts, each corresponding to one ofthe main sections ofthe technical report. Also number each of your graphs accordingly to be able to easily referto them, as necessary, in Sections 1, 2, and 3. You do not need to repeat graphs in Appendix B. For example, if you want to refer to a graph under the ARIMA section, which was already done in the section dedicated to exponentialsmoothing, you are encouraged to instead use the figure numberofthat specific graph rather than repeating the graph again.

2.4. Organizing your technical report. The report must be organized asfollows:

1. Exponential smoothing (maximal length: 2 pages)

Marksto beattributed based on how well you articulate the following aspects:

● Describe data preparation (and its effects) prior to the implementation of exponential smoothing methods.

● Describe preliminary analysis undertaken (and conclusions drawn) prior to the implementation of exponential smoothing methods.

● Give details of how exponentialsmoothing models were selected for each ofthe time series, and how effective these methods are at forecasting.

● Clarity and quality of presentation.

● Functionality of Python codes.

● Quality and suitability of illustrative or forecast result graphs.

2. ARIMA forecasting (maximal length: 1 page; total marks)

Marksto beattributed based on how well you articulate the following aspects:

● Describe any data preparation prior to ARIMA, and its effects.

● Describe preliminary analysis undertaken priorto ARIMA modelling, and the conclusions drawn.

● Give details of how an ARIMA modelwasselected,tested, and its effectiveness evaluated.

● Compare ARIMA and exponential smoothing forecasting, both in generalterms and in this particularinstance.

● Clarity and quality of presentation.

● Functionality of Python codes.

● Quality and suitability of illustrative or forecast result graphs.

3. Regression prediction (maximal length: 1 page)

Marksto beattributed based on how well you articulate the following aspects:

● Describe any data preparation prior to regression.

● Describe any preliminary analysis undertaken prior to regression and the conclusions drawn.

● Give details of how a regression model has been selected and comment on its suitability for prediction.

● Clarity and quality of presentation.

● Functionality of Python codes.

● Quality and suitability of illustrative or forecastresult graphs.

Appendix A: Description of your codes (maximal length: 1 page)

Marks here will be attributed based on how clear, informative, and brief is your description of what each of yourPython codes(or excel file, in case any preliminary operationsis carried out there) does.

Appendix B: Analysis and forecastgraphs (maximal length: 10 pages)

This appendix should beorganised in 3 sections, with the first,second, and third one dedicated to graphsrelated to the exponentialsmoothing, ARIMA, and regression methods,respectively.