Uni Stuff > Monash University
Monash University - Subject Reviews & Ratings
LifeisaConstantStruggle:
Subject Code/Name: ETC4420 – Microeconometrics
Workload:
1x 3 hour lecture per week
Assessment:
2x STATA exercises (20% in total) – Some questions and replicating some models on STATA – not very hard but quite lengthy
1x Assignment (20%) – Includes writing a report after estimating a few models taught in the lectures – not very hard as well, but very lengthy and hard to finish, especially for those who haven’t used STATA much.
1x Final exam (60%) – Students thought it was quite lengthy (I thought it was alright tbh), but aside from that, not the hardest exam ever.
Recorded Lectures: Yes, with screen capture
Past exams available: One
Textbook Recommendation: Econometric Analysis – William Greene – only 3 (very long) chapters are needed for this unit, specifically:
Part IV:
17 – Binary Outcomes and Discrete Choices,
18 – Multinomial Choices and Event Counts,
19 – Limited Variables, Truncation, Censoring and Sample Selection.
I thought the textbooks were useful in adding detail and mathematical explanations to the lecture slides, but it’s very long-winded and boring to read through.
Lecturer(s): Professor Xueyan Zhao – she’s very nice, and genuinely cares about her students. I’m more of a self-learner, and I thought I needed the textbook to supplement her lecture slides. Her explanations were a bit long-winded but that’s okay
Year & Semester of completion: S1 2021
Rating: 3.5 out of 5
Your Mark/Grade: 92 HD
Comments:
This unit is by and large an extension of ETC3410 – Applied Econometrics, with more models, and more weird things to consider in applied economic modelling and practice. We investigate extensions of the models in ETC3410 (binary, instrumental variables, panel data) to imperfect datasets, which are listed below:
1) Discrete choice – when the data is one of a few choices, e.g. ratings on Amazon (which is ordered), or preference in transport (which is unordered) – we look into multinomial choice models such as the ordered probit, multinomial logit, some of their extensions.
2) Counts – when the data is built based on a count random variable (e.g. number of hospital visits) – we look into count regression models such as Poisson or negative binomial models – and some extensions.
3) Data from non-representative samples (censoring, truncation, limited samples) – say for example, we only collected wage information from employed individuals, or recorded any wage below $X as <$X. We use the Tobit model, Heckman sample selection models (which are extensions of instrumental variable estimation in ETC3410) and associated extensions to deal with these defects.
4) Binary panel data – where the dependent variable is a binary choice (e.g. employment status of an individual across 2013-2021). Not too fancy.
Alongside the above we also have concepts such as efficient estimation, set identification, etc. that’s driving current econometric research where we have no definite answers to.
The unit is called microeconometrics because the models taught are usually specifically applied to microeconomic data – data that concerns individual units (people, businesses, and other microeconomic individual units), and how variables have causal effects on one another. Personally, I’m not too interested in these topics but I thought the unit was quite good in introducing advanced, open questions in econometrics that I haven’t thought of. Combining the conceptual, economics knowledge in this unit with more statistics/machine learning concepts and you’ll be a well-rounded practitioner of data analytics and modelling.
I thought the underlying philosophy of this unit is conveyed quite well – in that real-life data is often quite shit, which makes our models usually quite shit, even with some level of sophistication, and improving these models for better outcomes, instead of piling on complexity and hope for the best is an active area of research right now. Whether this unit is taught well is very subjective, but I thought it was alright.
LifeisaConstantStruggle:
Subject Code/Name: ETC4541 – Bayesian Time Series Econometrics
Workload:
2x 1.5 hour lectures per week
3x 2 hour tutorials the entire semester
Assessment:
3x Assignments (39%) – Questions pertaining towards the lecture slides, alongside some coding and simulation tasks. Quite typical of an ordinary econometrics unit. Not too hard but not easy too.
2x Learning Diaries (6%) – Just a record of what you’ve learnt and some reflections, pretty easy task.
1x Final Exam (55%) – Exam was quite lengthy, found it hard to finish with good answers but I thought it wasn’t the hardest paper ever.
Recorded Lectures: Yes, with screen capture
Past exams available: Several
Textbook Recommendation: Introduction to Bayesian Econometrics – Greenberg – Gives a very brief overview of each item we discuss in the unit, but not very useful beyond that. There are also some non-time series econometrics stuff that we don’t go through, and it’s quite interesting to read especially when you do the unit alongside ETC4420 - Microeconometrics, which teaches the frequentist equivalent of some of the equivalent models.
Lecturer(s): Associate Professor Catherine Forbes – a lot of people don’t like her ETC2420 unit, which she doesn’t teach anymore. But I thought she was quite good in this unit, in terms of answering questions and providing support as well. Definitely way better than ETC2420.
Year & Semester of completion: S1 2021
Rating: 4 out of 5
Your Mark/Grade: 93 HD
Comments:
I think a lot of people might have read about Bayesian statistics, but are usually unclear on how to explain them. It’s not a crazy awesome statistical paradigm that everyone should follow (otherwise people would be disappointed) – but it’s quite useful nonetheless.
Bayesian statistical inference is different by inference proper, which, aside from producing estimates, include producing quantities that specify our uncertainty on the estimate itself. In traditional, frequentist methods, we quantify this uncertainty using a sampling distribution (e.g. of the mean), which is usually only true in large samples, and cut the sampling distribution to report sampling uncertainty. Instead of a sampling distribution, we define all our estimates in a Bayesian case as probability density functions, which holds true in finite samples as well. Alongside some cool properties like Bayesian updating, dimension reduction, that’s basically the crux of Bayesian statistics and econometrics. Sometimes these are cool but it’s very hard to apply stuff in empirical work, and a lot of knowledge in computing and simulations are required to do these things quite well.
This unit is an introduction to Bayesian statistics, seriously, and applications of it to econometrics/time series, specifically, we look at:
1) Bayesian statistics in general – specifying priors, likelihoods, posteriors, and associated properties, Bayesian hypothesis testing, and complications when dealing with high-dimensional integration.
2) Simulation in Bayesian statistics – independent, Gibbs sampling, Metropolis-Hastings algorithms – which are important because we need to calculate and simulate values for many integrals in Bayesian statistics.
3) Bayesian econometrics and time-series – we investigate linear regression in a Bayesian setting, and state-space models (nested in a larger Hidden Markov modelling literature) within the Bayesian framework.
One can estimate state-space models if they assume a linear-Gaussian structure (LGSSMs), but if assumptions are relaxed, we can then only do approximate estimates (NGSSMs).
These things are quite cool and can only be appreciated once you do more time series stuff at a graduate level. Applications of these include factor models in macroeconomics – large models that aggregate macroeconomic variables for estimation, or even in electrical engineering/neuroscience.
If you are generally interested in time series, this unit, alongside ETC4410 – Macroeconometrics/ETF5200 – Applied time series econometrics are must dos at an honours level. Though the items taught in this unit are largely theoretical, their application in fields beyond econometrics are practically endless. State-space modelling in particular can extend beyond many concepts in the natural sciences and engineering, and skills like these are definitely transferrable and a rare find among undergraduate level topics.
LifeisaConstantStruggle:
Subject Code/Name: MTH5210 – Stochastic Calculus and Mathematical FInance
Workload:
2x 1 hour lectures
1x 1 hour tutorial
Assessment:
3x Assignments (30% - not sure) – I thought it were quite typical questions, and not too hard as well.
10x Weekly homework (10% - not sure) – Same as the above, but considering it was weekly, it was quite easy.
1x Final exam (60%) – As far as maths exams go, I thought the exam was very hard, and really stretches your knowledge. Knowing the textbook is quite essential.
Recorded Lectures: Yes, with screen capture
Past exams available: Several
Textbook Recommendation: Introduction to Stochastic Calculus with Applications – Klebaner – definitely a very important text because the lecturer is the author of this textbook, and he uses the content within the textbook very heavily instead of using slides or course notes (more typical for undergraduate Students)
Lecturer(s): Professor Fima Klebaner – he’s a very good and clear lecturer, clearly very experienced in teaching this unit (he’s also the author of the prescribed textbook), sets easy assignments too (the exam was a different story but it scales well)
Year & Semester of completion: S1 2021
Rating: 4 out of 5
Your Mark/Grade: 98 HD
Comments:
This unit is very much like MTH3251 – Financial Mathematics, but with more of a focus on stochastic calculus and underlying theorems. I guess it can be seen as a self-contained course in stochastic calculus alongside a bit of financial maths if you have some basics in calculus and probability. In particular, we look at:
1) Properties of Brownian motion (the most basic continuous time stochastic process – which are random variables that move through time)
2) Ito calculus – calculus incorporating Brownian motion (Ito’s formula underlies most, if not all studies in stochastic calculus – which is taught here)
3) Stochastic differential equations – diffusion processes (applications of Ito processes in ODEs to make SDEs and PDEs – in particular, we look at conditions for which strong solutions exist for a particular SDE, and how to find them – solutions are stochastic processes that have an explicit form that incorporate Brownian motion as well, we also talk a bit about martingales in more detail, which is relevant for a lot of theoretical work)
4) Weak solutions to SDEs – there are two ways of estimating weak solutions when strong solutions are not available, change of time, and change of probability measure, for which we investigate them here. The crux of it is just to define a new Brownian motion that makes the SDE solvable explicitly.
5) Applications to finance – we go through this very briefly, mainly discussion on applying the above concepts on asset pricing theorems and option pricing. These are all quite basic, but interested students can extend these concepts from these basics to complex methods – such as more complex derivative pricing, and modelling. (Applications are taught in MTH5520 – interest rate modelling where we look at stochastic calculus applied in bond markets).
I would recommend this unit compared to MTH3251, although the undergrad unit helps a lot with making this unit bearable. This unit goes much deeper into stochastic processes in continuous time and can be very interesting once you know where and why the mathematics is used in real-life applications – something the maths department doesn’t do very well tbh. The unit that is a straightforward extension to this is MTH5520 – Interest Rate Modelling, which uses stochastic calculus as well, and I find that doing MTH5210 makes the content there more bearable too, compared to a lot of students who only did MTH3251. The concepts stick easier, and it’s much more interesting that way.
Owlbird83:
Subject Code/Name: PHY2032 - Human physiology: Hormonal and digestive systems
Workload:
- 2h workshops weekly (except 3 weeks are 3h labs). Compulsory to attend or will miss 1% MCQ assessment.
- (1) to 2 online modules per week, each module contains ~30mins of lecture videos and questions. (takes ~6h to learn each week of content)
Assessment: (Outline the various assessments which make up the subject and how much each counts for)
- 10x 1% group MCQ tests (IFATS) during workshops (easy to get full marks or high 90s as you have a whole group and questions are fine if you've done the content).
- 10% Infographic in a group. With the group you work with in workshops the whole sem so need to make a good impression.
- 10% Stress report. Title and abstract based on a lab.
- 15% Hormone multimedia presentation. 6 min video providing info on a hormone (not chosen).
- 10% Neuronal control of gastrointestinal smooth muscle report. Based on a lab. Creating a figure legend and title for a graph, then answering questions.
- 5% SAQ assessment. Chose from 1 of 2 questions to answer under exam conditions (timed 20mins and open book). Very useful practice for the exam as it's the same question style.
- 40% end of sem exam. 2h 10mins. 30MCQs & 6 SAQs. (10 SAQs are given but only answer 6 (spread on each topic)). Open book using a lockdown browser.
Recorded Lectures: Yes. (But workshops are not recorded.)
Past exams available: Given 100+ practice MCQ questions on moodle, as well as 8 SAQs with videos explaining answers (+ some more SAQ practice without answers).
Textbook Recommendation: "Vander's Human Physiology" was recommended reading, however I didn't use it and don't think it is necessary as the lectures are very dense and contain all the info you need.
Lecturer(s):
A/Prof Craig Harrison
Dr Mike Leung
A/Prof Julia Choate
Dr Belinda Henry
Year & Semester of completion: 2021
Rating: 5 out of 5 (best unit i've ever done!)
Your Mark/Grade: will update
Comments: Give your overall opinion of the subject, lecturers, assessment etc. and a recommendation, plus anything else which you feel is relevant.
-Lecturers went above and beyond what most do. They put so much effort into creating workshops what were both fun and very useful. Each workshop contains some sort of video Mike and Craig are acting in and you need to apply your knowledge and answer questions as the storyline progresses, definitely makes it fun. Probably the most useful workshops i've had because they require you to work in the same group over the weeks so there's no wasting time being awkward and worried about others judging you at the start of each class. Also the questions you need to work on as a group are very useful for actually consolidating the knowledge and similar to the exam style questions.
-Lecturers were very organised and responsive on the forums to student questions and also understanding about adapting assignments to what is achievable without the in person lab classes. You could tell they put in a lot of work to transition from in person to online teaching.
-Because each 1% IFAT test is done in your workshop group, it forces you to keep up to date with the weekly content because you don't want your group to dislike you, which I found very useful so never fell behind.
-The weekly lecture videos are very content heavy. I found each 10 min video took around 40mins to 1h to write notes for because everything mentioned is so concise and important. Helps a lot to draw out all the diagrams of the anatomy and flowcharts for the processes, useful when you need to recall or refer to them in the workshops.
-4 topics covered: Endocrinology (wks 1-4), Reproduction (wks 5-7), Digestion (wks 8-9), Metabolism (wks 10-11).
-Cohort averages for assignments ranged from mid to high 70s, so felt like if you write according to assessment criteria you can get a decent mark.
-The assignment feedback was extremely detailed and useful. Also didn't feel like generic feedback.
-Overall, highly recommend doing this unit not only because of the interesting content but because of how well run it was and how good the lecturers were.
LifeisaConstantStruggle:
Subject Code/Name: ETC4410/5441 - Macroeconometrics
Workload: 1x 3-hour lecture weekly
Assessment:
1x 40% - Replication Assignment
Replicating a published paper in the macroeconomics literature, alongside some discussion and extension of results. This is quite interesting and accessible, albeit not being extremely easy. Also definitely a great plus to research and coding skills.
1x 10% - Literature Review
Providing a review for a working paper in the macroeconomics domain, which, I feel is quite interesting as well.
1x 50% - Final Exam
Recorded Lectures: Yes, with screen capture
Past exams available: Yes, 1
Textbook Recommendation: Lecture slides, recordings and MATLAB code are sufficient material, and there is no need for textbooks. The only book I would recommend for a deeper understanding would be James Hamilton’s Time Series Analysis, which is a graduate level econometrics textbook for PhD students, and goes into time series in way greater detail.
Lecturer(s): Benjamin Wong. Nice guy, one of the greatest and most insightful lecturers in the department. Not too chill with the content of the unit, which I feel is a good thing.
Year & Semester of completion: S2 2021
Rating: 5 out of 5
Your Mark/Grade: 91
Comments:
This unit mainly concerns the econometric methods used in macroeconomic modelling, particularly within research departments of central banks, and to a lesser extent policy research institutes. Theoretically it is concerned with 1. time series modelling of aggregate (macroeconomic) variables, stuff like GDP, inflation, oil prices etc. and their movement over time, and 2. Assessing the causal effect of changes/shocks in policy and the macroeconomic environment through econometric and statistical tools. You will cover a wide range of applicable models, a large chunk of literature and definitely more comprehensive than most econometrics units:
1. Univariate time series modelling
a. Autoregressive and moving average models (ARMA)
b. Forecasting and forecast evaluation
c. Impulse response functions
d. Bootstrap and local projection methods
e. Structural breaks
f. Simultaneity – and the use of instrumental variables in macroeconomics
2. Multivariate time series modelling
a. Vector Autoregressions (VARs)
i. Computation
ii. Forecasting, and conditional forecasting
iii. Bayesian VARs
b. Structural Vector Autogressions (SVARs)
i. Identification schemes to assess causal effects
1. Short-run/recursive/Cholesky’s identification
2. Long-run identification
3. Sign identification
4. Heteroskedasticity identification
5. External instruments
ii. Application of SVARs
1. Impulse response functions
2. Historical decomposition
3. Forecast error variance decomposition
c. Trend-cycle decompositions – decomposing time series into a trend, I(1) component and a cyclical I(0) component
i. Deterministic trends
ii. Hodrick-Prescott filter
iii. Bandpass filter
iv. Beveridge-Nelson decomposition
d. Factor modelling
i. The use of dynamic factor models
ii. Principal components analysis
e. Models for large datasets
i. Factor Augmented VARs
ii. Large Bayesian VARs
iii. Global VARs
iv. Mixed-Frequency VARs
f. State-space models
I thought this unit was really good for a few reasons, but it definitely wasn’t the easiest unit to score. This is probably the only unit that is directly related to the parent discipline of econometrics (economics), and teaches econometrics in a way that it is applicable to assess, understand, and solve macroeconomic problems. The content taught in this course is very comprehensive as well, bar some non-linear models and Bayesian statistics used for DSGE models, it covers pretty much most of what one needs to know to start off as a competent graduate at a central bank, or a research-centric job in macroeconomic consulting. One thing I particularly liked about this unit is that its assessments are very literature centric as well, and emphasises on bringing students to a level where they will be able to read more on, and understand the economics literature with greater ease (something not done in many other units).
With that being said this unit’s definitely not easy, and some might feel overwhelmed depending on their level and interest in macroeconomics and mathematics in general. I would recommend this unit only if you’re interested in applied macroeconomics (monetary policy in particular would be your best bet), and have done ETC3450/ETF5520.
At this level, macroeconomic modelling packages and libraries aren’t as well-developed compared to more popular statistical tools in software such as R or MATLAB. So coding up for the assignment can be a bit of a hassle, Ben provides MATLAB code, but the assignment can be done in EViews/Stata where macroeconomic modelling is rather complete in these software (a lot of people like R, but all I can say is that R isn’t particularly great for some purposes).
Anyways, this is a very good unit, I would definitely recommend it.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version