Assessment

Assessment Overview #

Assessment structure #

Your assessment for this subject comprises the following:

  1. a 60 minute closed book mid-semester exam (15%)
  2. an individual assignment (25%)
  3. a 2-in-3 hour online final exam (60%)

Hurdle requirement: To pass this subject students must pass the end of semester examination.

In the following sections, details are provided (and will be populated throughout semester) about those assessments.

Examinations:

Canvas Quiz will automatically submit the exam when the time limit (comprised of reading time & writing time) has been reached (for example, after 60 minutes for an exam with 30 minutes of writing time and 30 minutes of reading time). Students must click ‘submit’ at the completion time if they did not start the exam at the scheduled commencement time to not incur any late penalty.

For example, a 60-minute exam scheduled at 3:00pm, but commenced at 3:15pm, will automatically submit at 4:15pm and the student will incur a late penalty. Students must click ‘submit’ at 4:00pm to not incur a late penalty. Students who are prevented from submitting on time or at all due to technical difficulties will need to apply for technical consideration with supporting documentation.

Mid-semester exam #

Date #

The mid-semester exam will take place online on Tuesday 23 April 2024 between 1.30pm and 2.35pm. Please arrive by 1.15pm at the latest. The exam won’t start before 1.30pm, but it may start later, so please make sure you are available until 4pm.

Type #

The mid-semester exam will be a “traditional”, closed book, pen on paper exam.

You will be allowed to bring one A4 sheet of notes, double sided. You can write as much as you want on that one sheet of paper (text and/or formulas), on both sides, but it cannot be more than one A4 sheet.

You will be allowed to bring a Casio FX-82 (with or without suffix) calculator.

Venue #

PAR-Kwong Lee Dow-101and102

Duration #

60 minutes sharp, plus 5 minutes reading time.

Scope #

The mid-semester exam will assess contents from the first half of the semester, that is, Modules 2 and 7 to 10 (lectures of weeks 1-6) and associated tutorial exercises, readings, R codes, and course contents. This focuses on the time series component of the subject, with some claims modelling as well.

Examinable materials include associated tutorial questions, revision questions, additional questions, technical and practice lectures with slides and associated mandatory readings.

Practice #

The exam will be of the “traditional” type, in line if the IFoA CS2A exams; see below for details.

In the community website under modules there is a “Pen-and-Paper Mid-semester Practice/Sample Exam”:

  • This exam was put together specifically for the first closed book exam in the history of ACTL30007/90020 and includes questions that were never used in this subject.
  • It was meant as a sample for Modules 1-5 for the 2023 edition of the subject.

Additional details #

See General Advice for further details.

Individual assignment #

Assignment instructions #

The assignment will be a data analysis assignment which will need to be performed in R, and you will need to submit a 5 minute oral presentation, slides, and R code.

The assignment will assess skills from Modules 7 to 9 (TBC) and associated tutorial exercises, readings, R codes, and course contents. This focuses on the time series component of the subject. Communication skills will also be assessed.

Additional details and instructions will be available later (before the Easter non-teaching period).

Assignment submission #

Assignment submission is via the LMS Assignment Submission link for all written assignments. Please refer to the Turnitin section of the LMS website via for detailed submission instructions if needed.

Please note that you are required to keep a copy of your assignment after it has been submitted as you must be able to produce a copy of your assignment at the request of teaching staff at any time after the submission due date.

Penalties for Late Submission #

In order to ensure equality for all students, assignments and examinations (where relevant) must be submitted by specified deadlines. Late assignments, where approval for late submission has not been given, will be penalised at the rate of 20% of the total mark per day, for up to 5 days, at which time a mark of zero will be given.

Students with a genuine and acceptable reason for not completing an assignment (or other assessment task), such as illness, can apply for special consideration. Special Consideration assists students who have been significantly affected by illness or other serious circumstances during the semester. The following website contains detailed information relating to who can apply for Special Consideration and the process for making an application: http://students.unimelb.edu.au/admin/special

Final exam #

Date #

The date of the final exam is published on the University of Melbourne Exams Timetable website.

At your scheduled date, the final exam will be available on the Canvas website under "Assignments".

Duration #

Important online examination information: This examination is the equivalent of a 2-hour assessment; however, students are provided with 3 hours to accommodate the time that will be required for typesetting of mathematical expressions, loading, processing and modelling of data with the statistical software R, and the uploading of documents / answers. There is also a 15 minute reading time, during which you may write and edit your answers already.

Exam conditions #

In 2024, the exam is open book but invigilated, and will take place in the Wilson hall, with a University-provided laptop.

Details are available in the following document, which outlines:

  • Exam setup: how to access your University-provided laptop, Canvas and authorised applications
  • How to store (prior to the exam) and access (during the exam) personal files
  • How to create a RStudio project on your University-provided laptop It is strongly recommended you practice this on your own laptop as this is the best way to minimise issues with file location in your R code
  • How to import your data sets
  • How to add images from R Studio into your Canvas quiz It is strongly recommended you practice this before the exam in the Practice quiz
  • How to navigate and submit your quiz
  • What extra support you have available during the exam

I strongly recommend you read this document, and save it in your files for easy access during the exam.

Scope #

The final exam will assess all contents of the course, with particular focus on the materials taught after week 6 (after the mid-semester exam). This means there will be questions on both the claims modelling and time series components of the subject.

Examinable materials include associated tutorial questions, revision questions, additional questions, technical and practice lectures with slides and associated mandatory readings, as well as the assignment task and the mid-semester exam.

Marks and allocation to topics #

The final exam will comprise 6 questions for a total of 60 marks.

Here are a detailed breakdown of the questions:

  • Per module:
    • 31 marks on M2-M4
    • 8 marks on M5
    • 8 marks on M6
    • 13 marks on M7-M10
  • Per type:
    • all questions are short essay questions (no MCQ and no MAQ)
    • about 2/3 of marks require the use of R. Only one question (out of the 6) does not involve R at all.

While you can answer everything on Canvas (and in some cases that will be mandatory as you will need to include some R output and images), you will have an option to answer some questions by hand. This is the case of questions 1 to 5, and the booklet is available (click on the link to see it).

Practice #

See Exam preparation as well as past subject exams and CS2 exams

Exam consultation #

Here are the final consultation times:

  • 29 May 3-5pm - in person FBE-210
  • 3 June 3-5pm - in person FBE-210
  • 7 June 3-5pm - in person FBE-210
  • 10 June 3-5pm - in person FBE-210
  • 11 June 2-4pm - in person FBE-210
  • 11 June 4-6pm - on Zoom

As previously announced, in-person consultations will also be available on Zoom for (virtual) drop-ins if no-one is around, but in-person visitors will have priority.

Zoom links will be available on Canvas.

Additional details #

See Exam Format and General Advice for further details.

See also the University of Melbourne exam webpage for students.

Format of online exam #

Types of quiz questions #

There will be three types of questions:

  1. Multiple Choice Question (MCQ) with exactly one correct answer. These are similar to those used in the mid-semester exam. Students get full mark for a correct answer, and 0 marks for an incorrect answer.
  • Example: What is the result of 2 times 4?
    1. 4
    2. 6
    3. 8
    4. 10
  • In this example there is only one correct answer – 3.
  1. Multiple Answers Questions (MAQ) In this case, there are multiple choices, but the number of correct answers can be different from 1 (it can be anything from 0 to the number of possible choices). Here is the explanation of Canvas about how marks are awarded: “To calculate scores for Multiple Answers quiz questions, Canvas divides the total points possible by the amount of correct answers for that question. This amount is awarded for every correct answer selected and deducted for every incorrect answer selected. No points are awarded or deducted for correct or incorrect answers that are not selected. For example, an instructor may create a Multiple Answer quiz question with 9 points possible that includes three correct choices and two incorrect choices. If a student selects two correct answers and one incorrect answer, they would be awarded 3 total points for that question. This would be calculated by awarding 3 points (9 total points divided by 3 correct answers) for each correct answer and subtracting 3 points for the incorrect answer.” Furthermore, there is a minimum score of 0 (you can’t get a negative score).
  • Example: Which of the following are negative?
    1. 4
    2. -6
    3. -8
    4. 10
  • In this example there are two correct answers: 2. and 3.. If there are 4 marks allocated to this question, marks would be awarded as follows:
      1. and 3.: 2+2=4 marks
      1. only: 2 marks
      1. and 4.: 2-2=0 marks
      1. and 2. and 4.: -2+2-2=-2 but there is a minimum score of 0, so 0 marks.
  • With this type of question you must carefully assess every single answer independently of the others.
  1. Essay Questions In this case, you must answer with text, and you get a text field to compose your answer. Note that formulas can be composed too. You may also upload a handwritten answer if you wish, although we recommend not to do this unless formulas are required extensively. These questions will be marked by someone as they cannot be automatically graded.

Precision of numerical answers #

Students often ask about the required significant figures / decimal places.

If it is not specified:

  • In an open question we will look at the reasoning so you should not be too worried about it (within reason). Use, say, 5 significant figures/digits ∗
  • Otherwise the precision needs to match that of the possible answers (if we have 5 significant figures ( ∗ ) in the answers then work with 6, say).
∗ Please move away from talking about decimal places - this does not take the scale of your problem into account. The correct way to think about this is in terms of significant figures/digits. For more details, see e.g. Wikipedia or this video from Khan academy.

Exam preparation #

Assessments in this course will generally:

  • test whether you can apply the methods in a realistic environment (in particular, with data and R);
  • test whether you understand the theory behind those methods, and can explain the associated strengths and limitations.

To prepare for this exam we suggest you review the tutorial exercises and past exam questions. You should also make sure you read all prescribed readings and have understood the main arguments. Remember that this is an open book exam, so there is nothing to memorise. You must demonstrate understanding instead, and being able to “connect the dots”. If you have done the work and know where the things are, you should be able to easily find the details you need to answer the questions.

More specifically:

  • Finalise your own summaries and formula sheets in preparation for the exam.
  • I recommend you review carefully the Detailed Learning Outcomes pages:
    • For Actuarial Practice: are you able to do all these? (describe, list, explain, …); if you are on top of those you should be well prepared for the final exam;
    • For Actuarial Techniques: do you understand all those formulas? Do you have formulas to add? Is there anything to add to your own summaries?
  • Of course, review also all tutorial exercises. For each question:
    • What skill(s) is this exercise testing?
    • Where are the required concepts in your summaries? in your formula sheets?
    • What did you find challenging in this exercise, and how comfortable are you with this now?

Also, have a look at the past exams below!

Specific preparation for R questions #

Here is general advice about how to prepare for that component of the subject:

  • After the lectures (or before), make sure you implement all codes on your own machine. These can be copy pasted from the website easily, and there is nothing that is missing. I am using those exact codes to produce the outputs; there is no hidden chunk.
  • Try to understand what each line is doing. Execute the code line by line, change parameters, see what it changes. Understand why it works or not.
  • For each function you don’t know read the help in R. For function foo, simply write ?foo in the console.
  • Google up issues, ask on Ed if you are stuck.

You can’t wait until the exam to work out the codes. In the mid-semester exam, one could essentially copy paste the codes from the slides. In the final exam you may have to adjust slightly one way or the other, and for this you need to be on top of the lecture slide code.

Being able to implement those methods is really important, and what makes the assessment “authentic”. This is how you would actually use that knowledge in practice, not via proving theorems (although that can be fun, too, but it is not too well suited for an open book exam).

General Advice #

Past subject exams #

In the community website under "Modules" you will find the following past subject exams:

Final Exam (2023) #

  • This is the complete exam from 2023
  • Covered all modules, but mid-semester was on Modules 2-4 and Assignment was on 2-4, so focused more on the other modules
  • Had about 1/3-1/2 of MCQ questions
  • Had about 1/2 of R questions

Mid-semester Exam (2023) #

  • This is the complete exam for 2023
  • Covered Modules 2 to 4

As this was a traditional pen on paper exam, it:

  • Had no MCQ questions
  • Had no R questions

Final Exam (2022) #

  • This is the complete exam from 2022
  • Covered all modules, but mid-semester was on Modules 2-3 and Assignment was on 7-8, so focused more on the other modules
  • Had about 1/3 of MCQ questions
  • Had about 1/3 of R questions

Mid-semester Exam (2022) #

  • This is the complete exam for 2022
  • Covered Modules 2 and 3 only
  • Had about 1/5 of MCQ questions
  • Had about 1/2 of R questions

Final Exam (2021) #

  • This is a partial exam (only 46/60 points) only, covering Modules 5-10
  • Note other Modules may not be representative as some materials were made out of scope after 2021, especially in Modules 1-4
  • Most of the questions removed were non-R and non-MCQ questions

Mid-semester exam (2021) #

  • This is the complete exam for 2021
  • Covered Modules 7-9 exclusively
  • About 1/4 were R and non MCQ questions, the rest were just MCQ questions

Final Exam (2020) #

  • This is an old format (from first Covid lockdown!), with just three versions of each question each time.
  • Take this as extra practice, but question type are more representative of closed book exams, than the more modern “open book with R” exams of more recent years.

Past CS2 exams #

This subject is one of three subjects leading to exemption from the IFoA CS2 exam; see also the SILO page. This provides with a very extensive range of practice exams. Note:

  • CS2A exams are closed book pen on paper exams
  • CS2B exams are R based
  • Past exams and solutions are downloadable here

A mapping of the relevant questions, into Modules of this subject, is available here.

Tips #

Remember there are different ways of answering a multiple choice/answer question (especially if it involves numbers):

  • Work out the answer as if you were in a standard pen and paper invigilated exam, then see if the answer is in the list.
  • Work out the answer with tools that would not otherwise be available, such as Excel spreadsheet, R, or even the internet. (but be careful not to collude or otherwise be guilty of academic misconduct – check https://academicintegrity.unimelb.edu.au/ for further details)
  • In some circumstances, it may be easier to simply “plug back” the different answers to see which one works. Don’t forget these three options exist for multiple choice/answer open book exams! Hence make sure you have Excel and R available and working before the exam starts.

Furthermore, some more advice / exam technique / tips:

  • Read attentively the exam questions at least twice before attempting the question. Work out what skills you are meant to demonstrate. Make sure you understand what is being asked in the question – consider key words indicating what you have to do. When you have finished, re-read the question to make sure you have not missed anything.
  • Key is to demonstrate understanding.
  • Attempt all questions where possible.
  • If you run out of time, explain how to you would approach the task (where possible). If this is right this should give you marks.
  • Do questions you are most confident with first. Prioritise.
  • Observe the number of marks allocated. Manage time accordingly. Reassess your available time regularly (say, half way through for a 1 hour exam, and perhaps 2 times for a 2 hour exam). For instance, if the exam is one hour and there are 100 marks, then you have 36 seconds per mark.

Pay close attention to the wording of the question - this is not chosen randomly. Pay particular attention to the verb (for instance: “State” and “Show” do not mean the same thing. With “State” you just need to give the answer, with “Show” you need to actually prove the answer) and adverb if any (for instance: “briefly” is not the same as “in detail”). Examples are:

  • ‘…Justify your answer’
  • ‘Find’
  • ‘Explain in words’
  • ‘Derive’
  • ‘Show that’
  • ‘Briefly list’
  • ‘Present in detail…’
  • ‘Briefly explain’
  • ‘Briefly describe’
  • ‘State (without justification) 4 examples of …’