ST5188程序代写、代写R，Python编程

日期：2023-04-06 08:51

ST5188 Statistical Research Project

(AY 2022/23, Semester 2)

– Course Guide –

Version 1.0

ST5188, which will be conducted in e-hybrid mode, is a project course. As such, there will

be no regular lectures or tutorials. Throughout the semester, students will work (in groups)

on their projects independently. However, there will be dedicated touch points with

facilitators including three lectures, consultation sessions (for each group) with the

lecturer, and TA support.

Given that ST5188 is a project course, it is evaluated as such. Different students may

contribute in different ways; according to their skills, abilities and project plan agreed upon

by the whole group. However, all students are expected to contribute similar efforts to the

project.

All the students in a group share equal responsibility for creating team spirit and making

the group work as a whole. Should a problem arise, each student must be willing to work

towards resolving the problem. Do not hesitate to ask your assigned TA for mediation;

should problems persist, the TA will escalate the matter to the lecturer.

ST5188 assessment components are as follows:

Contribution Due Date

Late Submission

(25% penalty applies)

Project Proposal 15% Feb 8th, 11:59pm up to 48 hours

Project Progress Report 5% Mar 15th, 11:59pm up to 48 hours

Project Presentation 15% Week 13 No

Peer Project Evaluation 5% Apr 16th, 11:59pm No

Final Project Report 60% Apr 16th, 11:59pm up to 48 hours

ST5188 will be graded on a completed satisfactorily / completed unsatisfactorily basis!

There will be three lectures (90mins each; conducted in LT26 and streamed via Zoom):

Jan 9th (Mon), 5pm: ST5188 Introductory Briefing Session

Jan 14th (Sat), 2pm: How to Write an Effective Capstone Project Proposal

Jan 28th (Sat), 2pm: Data Science Project Best Practises

For course details and updates, please refer to the ST5188 Canvas page.

2 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

ST5188 Summary

Students will have an opportunity to conduct basic research activities on a topic of interest.

The topics are either self-proposed, proposed by a faculty member, or real-world problems

from companies, research institutes, etc. Students will work on projects in groups of three

students (default). However, for real-world problems sponsored by companies, research

institutes, etc. smaller group sizes are permitted, but subject to approval by the lecturer.

During the course, each group of students will progress through three stages:

1) Formalise a project proposal (typically, weeks 2 to 4);

2) Conduct basic research activities (typically, weeks 5 to 13); and

3) Present their approach and findings (typically, week 13).

Throughout the semester, project groups will have the opportunity to get direct feedback on

their project progression via consultation sessions.

Learning Outcomes

A student will acquire many of the following skills:

Conduct systematically an extensive literature search and review of a topic of interest.

Review critically the papers read.

Propose feasible research projects arising from a literature review.

Perform coding in R or Python.

Write a report summarizing all the findings.

Present the findings in an oral presentation.

Discuss the findings with classmates and the lecturer / TAs.

Pre-Requisite

ST5201/ST5201X Statistical Foundations of Data Science; and

ST5202/ST5202X Applied Regression Analysis

Modular Credit

4

Workload

0-0-0-8-2

Teaching Modes

E-hybrid mode.

This is a project course; there will be no regular lectures / tutorials. Throughout the semester,

students will work on their projects independently with the lecturer doubling as consultant.

Grading Basis

Completed Satisfactorily / Completed Unsatisfactorily (CS / CU).

CS means scoring 50% or above; a CU grade will be awarded for scoring below 45%;

all scores in the range of 45—50% will be adjudged on a case-by-case basis.

3 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

What to Expect: Example of Typical Activities (week-by-week)

Students are given the freedom to choose their own project groups, project topics, and

approach to work towards tackling the project. However, with such freedom comes greater

responsibility, accountability, and communication requirements (with your peers and

teaching staff). Below, you find a sample timeline of core activities and course deliverables.

Week Core Activities (Example) Deliverables (Example)

1  Read up on ST5188

Attend ST5188 introductory briefing session

Form and register project group

Attend lecture on “How to write an effective

capstone project proposal”

2  Explore project topics

Literature review (wrt. project topics explored)

3  Formulate project scope, objectives / deliverables,

approach, success measures, and project plan

Schedule first consultation with lecturer

Attend lecture on “Data science project best

practises”

4  Attend first consultation session with lecturer

Finalise project proposal

Commence project work

5  Work on project ? Project Proposal

6  Work on project

Schedule second consultation with lecturer

Recess Week

7  Work on project

Attend second consultation session with lecturer

Complete first peer group evaluation form

Peer Group Evaluation 1

8  Work on project

Write project progress report

9  Work on project ? Project Progress Report

10  Work on project

Schedule third consultation with lecturer

11  Work on project

Attend third consultation session with lecturer

12  Work on project

Plan for project presentation

13  Present your project to lecturer, TAs, and peers

Complete work on project

Attend two peer project presentations and write

peer project evaluations

Finalise final project report

Complete second peer group evaluation form

Project Presentation

Peer Project Evaluations

Project Report Submission

Peer Group Evaluation 2

14 Reading Week

15 Examination Week 1

16 Examination Week 2

4 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

Getting in Touch / Asking for Help / Receiving Feedback

Throughout group project activities, your lecturer will double up as your group’s consultant

and assist with brainstorming and decision-making activities; plan ahead to ensure that you

use this opportunity wisely. Consultation sessions (20 minutes each) commence in week 3

and are available until the end of week 12 (each group may book up to 3 such consultation

sessions). In addition, each group will be assigned a teaching assistant (TA) who should be

your first point of contact wrt. any queries related to the project or the course in general. If

and as necessary, TAs will escalate enquiries to the lecturer.

Besides consultation sessions and TA support, you will also receive detailed written feedback

for your first two assignments, your project proposal as well as your progress report, and

verbal feedback during the Q&A session of your group’s presentation during week 13.

Recommendations for consultation sessions

You are recommended to arrange for up to three consultation sessions with your lecturer;

the first such session MUST be completed before submitting your project proposal.

You can either book an online consultation session (conducted via Microsoft Teams)

or an in-person consultation session (held in Blk S16, 6 Science Drive 2, Singapore

117546); please choose the mode that is most appropriate for all your group

members. A booking form will be made available via the ST5188 Canvas page.

For online consultation sessions, group leads should ensure that the meeting link is

circulated to all group members well in advance of the agreed upon consultation slot.

Consultation sessions are meant for you to get feedback and/or inputs from the

lecturer. As such, you should have an agenda on what topic(s) to discuss / question(s)

to raise. Ideally, you will run consultation sessions; your lecturer will then add-on and

keep an eye on the time.

Since every group will work on a different project, you will need to set the context. For

this, we strongly suggest that you send a summary (limited to half a page or two

paragraphs) of your project idea, project progress, key discussion points or else (at

least 2 hours) PRIOR to the consultation session. That way, the lecturer can have a

pre-read, and you have the full 20 minutes of your session available for discussion.

Be ready a few minutes BEFORE the start of your session. If some or all your group

members join / arrive late, you will lose valuable discussion time.

5 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

Key Activity 1: Project Formulation & Project Proposal

ST5188 group projects will require you to apply current or emerging statistical concepts,

methods or techniques to an interesting application or real-world data sets. Each project

should include some form of analysis and some form of experimentation on real-world or

synthetic data sets.

Throughout group project activities, your lecturer will double up as your group’s consultant

and assist with brainstorming and decision-making activities. Each group MUST schedule and

attend their first consultation session with the lecturer prior to submitting the project

proposal.

The first key activity is comprised of two parts:

1) In-depth review / study of a statistical concept, method, or technique that falls under

the scope of the MSc (Statistics) by Coursework Programme:

Statistical foundations of data science, applied regression, design of

experiments for product design and process improvements, nonparametric

regression, analytics for quality control and productivity improvements,

analysis of time-series data, multivariate data analysis, sampling from finite

populations, survival analysis, categorical data analysis, probability and

stochastic processes, statistical analysis of networks, spatial statistics, and data

mining.

2) Proposal of a group project related to your concept study from part 1).

Guidelines for part 1: In-depth review / study of a statistical concept, method, or technique

You are expected to learn BEYOND what was covered in prior coursework. You may think of

it as a literature review comprising of a technical concept summary, concept critique (pros

vs. cons, limitations wrt. assumptions made, practical feasibility or scalability of approach,

etc.), and initial brainstorming of at least two potential project opportunities (e.g., open

questions, better model, better algorithm, test on different data set, reformulation / removal

of assumptions, apply to different domain, etc.). Based on this, you will then work on the

second part, your group’s project proposal.

6 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

Guidelines for part 2: Proposal of a group project

Your project proposal (up to 6 pages including a 2–3-page literature review / concept study)

should cover the following subjects:

Project title (this can be a working title)

Project introduction / motivation

Problem statement or hypothesis

Literature review / concepts study (2-3 pages)

Project objective(s)

Requirements (in terms of data sets [how to obtain / generate them], compute

resources, tools, etc.)

Success measure(s) (in terms of evaluation, experimentation, testing, etc.)

Project plan including key activities

o Note: Take into consideration the size of your group and time allotted for this

group project.

The second lecture (to be conducted on Jan 14th at 2pm) will focus on how to write an

effective project proposal. Please carefully review the lecture to ensure that you meet

expectations!

Project proposals are due by Feb 8th, 11:59pm. You are highly recommended to submit your

project proposals early (i.e., during week 4). We aim to return detailed project proposal

feedback five to eight working days after the respective proposal has been received. That is,

the earlier your submission is received, the earlier you will receive corresponding feedback!

Project proposal templates as well as submission details will be made available via the ST5188

Canvas page. You are only required to submit one project proposal per group.

7 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

Key Activity 2: Project Progress Report

As a rough guideline, your group should have completed at least 50% of the work by the time

you submit your progress report. That being said, there may be circumstances under which

this is not feasible (e.g., in case your group decided to significantly revise the project’s focus

or objectives). However, such circumstances should be made known to your assigned TA /

lecturer as soon as they arise and not only during progress report submission. Thus, please

make sure that any significant project proposal amendments have been discussed during a

consultation session prior to project progress report submission.

In the project progress report, you should provide a summary of your progress thus far (i.e.,

demonstrate that you have completed 50% of the work) as well as a discussion of any

unforeseen challenges or difficulties encountered (you can limit yourself to discussing three

such challenges / difficulties). This report gives each group also a chance to re-evaluate initial

problem statements, objectives, and contributions (espc. for those groups who have been

recommended to revise such sections in the project proposal feedback) and make

amendments to the project scope / plan.

Your project progress report (up to 3 pages) should cover the following subjects:

- Project title

- Discussion of project progress (e.g., what has and what has not yet been finished;

difficulties / challenges encountered; and changes made to the initial problem

statement, objectives, assumptions, success measures, and/or contributions)

o We strongly encourage you to include an explorative data analysis (to

demonstrate that your chosen data-set(s) is/are suitable for your project

scope) unless you have already done so in your Project Proposal submission.

This also forms a good baseline for the discussion of challenges and difficulties

encountered (as they are often data pre-processing related in the early stages

of project work).

- Updated project plan for remaining / unfinished work items

Project progress reports are due by Mar 15th, 11:59pm. You are highly recommended to

submit your project progress reports early (i.e., during week 8). We aim to return project

progress report feedback five to eight working days after the respective report has been

received. That is, the earlier your submission is received, the earlier you will receive

corresponding feedback!

As a general guideline, you are encouraged to have completed your second consultation

session with the lecturer prior to submitting your progress report.

Project progress report templates as well as submission details will be made available via the

ST5188 Canvas page. You are only required to submit one project progress report per group.

8 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

Key Activity 3: Project Presentation

Each team will be given a 30-minutes presentation slot during week 13 – corresponding

details such as venue and sign-up sheets will be announced via the ST5188 Canvas page. All

team members are required to attend. Presentations are time-constrained (please plan for

about 15 minutes of presentation time followed by 10 minutes of Q&A; we will reserve 5

minutes for set up and other contingencies). During your presentation time (about 15

minutes), we will not interrupt for questions; however, we will stop you after 20 minutes

should you exceed the presentation guidance by more than 5 minutes and move on to Q&A.

For the group presentation, there is no prescribed format; you may use slides, but you don’t

have to, you may ‘walk through’ your project deliverables, or else. You are not expected to

cover every aspect of your project during the presentation (there is not enough time for that);

instead, your presentation should give the audience an overview of what your project is

about, highlight notable challenges and/or achievements, and, ideally, motivate the

members of the audience to have a desire to read your final project report.

Each presentation will be attended by all members from two other project groups, one of the

TAs (but not necessarily the TA who was assigned to your group as primary point of contact),

as well as the lecturer. As such, please do not assume that the members of the audience are

familiar with your project; you will have to provide corresponding context at the beginning of

your presentation.

During the Q&A session, questions may be asked by any of the members in attendance.

9 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

Key Activity 4: Peer Evaluation

Each student will have to participate in two types of peer evaluation activities:

1) Peer group evaluations: At the beginning of week 7 and after the final project report

submission, each individual student will be asked to provide feedback on the

performance / contribution of his/her team members as well as a self-evaluation.

Peer group evaluation feedback may be taken into consideration should there be

significant differences in contribution by the group members.

2) Peer project evaluation: During week 13, each group will be asked to (virtually) attend

two other group project presentations; all group members are required to attend. The

attending groups of students will be asked to complete a feedback / evaluation form

(e.g., provide a brief summary, three key achievements, and three constructive

feedback items) and are strongly encouraged to ask questions – but should do so only

during the allocated Q&A slot after the corresponding presentation has completed.

Peer evaluation / feedback forms will be made available via the ST5188 Canvas page.

10 | P a g e M a r k u s K i r c h b e r g

Dept of Statistics and Data Science :: Faculty of Science :: National University of Singapore

Key Activity 5: Final Project Report

The final project report must not exceed 10 pages (using the final report template, which will

be made available via the ST5188 Canvas page). It should be a concise reflection of your

group’s project work; content may cover the following aspects:

- Project title

- Abstract

- Detailed problem description (including context, problem statement, objectives, and

assumptions)

- Related work

- Description of data collection and pre-processing processes

- Description of the concept, method, model, algorithm, and/or technique that is/are

at the core of your project

- Discussion of key challenges

- Evaluation of your experiments / results / findings (including lessons learnt)

- Recommendations for next steps or follow-on efforts

- Outline of each team member’s contributions

- List of references (using either APA version 6 or 7, Chicago or IEEE formatted references).

Your final report should be self-contained (i.e., do not assume that the reader is familiar with

any of your prior submission documents). We will evaluate and assess the final report as an

individual / stand-alone document.

While there is a 10-page limitation, you may add supplementary details (e.g., a full description

of all the variables covered in your dataset, additional or enlarged versions of tables or figures

that are hard to read in the main body) in the appendix. In addition, your reference list may

extend beyond page 10 as long as the first reference appears on page 10 itself.

In addition, you will be asked to submit your code along with a documentation detailing how

to replicate your work.

Final project reports are due by Apr 16th, 11:59pm. Submission details will be made available

via the ST5188 Canvas page. You are only required to submit one final project report per

group.

Note: Final report submissions will be examined for plagiarism! You are strongly encouraged

to self-inspect your final report documents prior to submission. As a general guidance, your

final report’s similarity score must be below 20% overall with no individual paper contributing

more than 2% to the overall score.

End of Course Guide – ST5188 Statistical Research Project