ETC3550: Applied forecasting

This has got to be the most well-run subject at Monash. Rob has so many resources at hand that you are unlikely to be lost for help when you are undertaking this subject.
The main advice I would give is have some background in R and RStudio before you take this subject. The subject uses R and RStudio regularly to replicate methods taught in the lectures. So, having some background in that will come in handy, as you will be comfortable with working in R on a regular basis.

- These were standard except that they were run on Zoom due to 2020 circumstances. You mainly worked on problems assigned each work, most often in Zoom breakout rooms to maximise discussion between students. The problems ranged from being relatively straightforward to somewhat difficult, particularly in the mathematical components. That being said, detailed solutions were provided to the problems, so they did make sense after going through the solutions. Make sure you go through them before the exam, as it will help with getting a deeper understanding of R coding and interpretation of real-life datasets. I had Sayani as my tutor who was excellent, but Rob has such a vast calibre of PhD students that all of them are amazing in their own right.

Other than that, this is a really fun and interesting unit to take. Rob has developed a perfect university course based on his own research, that when learning the unit you really feel like you are experiencing the brains behind all forecasting methods used at Monash. Plus the unit is that well recognised that there is an award here that awards a US$100 prize, a Certificate of Achievement, and one year’s free membership of the International Institute of Forecasters (IIF) to the highest scoring student in the subject. So, if you do well enough, you become an automatic member of the IIF, which is amazing as it translates work done in university into practice. Do not miss out on the opportunity to take this unit if you can, you will not regret taking it.

. This was delivered as a Moodle Quiz for the first time. It followed the same structure as previous exams, but included some coding in R, with students being required to paste output into Moodle. Due to this, 3.5hrs + 10 mins reading time was allocated to ensure that everyone had enough time to write down answers and upload them, or copy code in R into Moodle. This was difficult to fully implement, and tough marking also meant that grades were lower than expected overall. Despite this, the exam was relatively achievable, due to it being open-book, but it did still present some challenges such as minor technical difficulties.

Yes, past 5 years of exams provided with solutions

- Rob is amazing. Please try and do this unit if you can. The amount of resources he has for this unit is astonishing. Rob has been teaching forecasting at Monash for over 15 years, and it shows. He has developed nearly all of the forecasting methods used, and they make understanding the content far easier when you are taught by someone who wrote everything being used in the subject.

5 out of 5

Yes, with screen capture

. This was mainly just taking everything you learnt in the unit and applying it to a real dataset. Each student was given a random dataset based on their student ID. This was taken from the ABS data on retail trade. From there you applied methods in class to forecast future sales, and then wrote that up in a neat report in R. This was not too hard to complete, but there are many tiny details that you can miss if you do not fully the rubric properly, such as graph labelling and selecting parameters for various tests. Make sure you check over every part of your project before you submit, as there are intricate details that you could be caught out on.

. Each week you are given a short assignment to do. Normally there would be 10 short assignments, each worth 2%. But this semester was a little different, with there being 8 assignments, 2 being worth 4% and the remainder worth 2%. Usually, these did not take too long to complete, but occasionally they could be slightly confusing. Particularly when it came to mathematical derivations, make sure you know all the content being taught and can apply it to ensure you get full marks in the in-semester assignments.

2 x 1hr lectures per week
1 x 1.5 hr computer laboratory per week

Semester 1, 2020