Post-Course Comments from 2007 SimSS Participants

Post-Course Comments from 2007 SimSS Participants

Dear Bob,

Here are some preliminary thoughts as regards the feedback to the course.

First of all, I think this is a very helpful course, at least to me. At the beginning, I actually didn't know what my expectation was. The topics were not frequently mentioned in other seminars I attended, nor were they an important part of my own thesis. But after the course, I actually do think this could be a very important knowledge building brick, which is at least as important as other methodological courses offered to PhD students.

Second, obstacles for students to attend may come from the unfamilarity or inconfidence of the topics. If they know they can overcome them, probably they would be more willing to attend.

Third, for the individual topics, I found Agent-based Simulation and Network Analysis were the most relevant models to me. It is also very good to know CA (as a lead for more complex models) and GA (the Szpiro [1997] paper gives us a good example of how we can relate it to researches in business). Relatively, System Dynamics is of less importance to business students.

Fourth, I not so sure whether a little bit of knowledge on programming should be provided, since I consider it a device to arouse interests. Probably at the beginning of each session, some NetLogo examples are shown and people can discuss for a few minutes on what are the aims of the modeler. You may also ask students to point out the parameters in the models.

Finally, I want to take this chance to say "thank you" - I never regreted sitting in the course :)

By the way, can I consult you on network analysis in the future?

All the best,

Dear Bob,

Thank you very much for your course on simulation!

My research area is about organizational design. After taking your course, I am keeping on thinking whether the simulation method can shed lights on my field. I think, if the organization can be considered as a complex system, there will be a great role for simulation to play. However, I find only a small number of articles used the simulation methods, such as ABM, to investigate the organizational design issue. It still eaves a large space for me to explore.

Another issue here is that I find the simulation is highly related to the complexity theory. The terms, such as landscape, fitness, and genetic, frequently appear in the simulation articles. Sometimes, I got confused during my reading. I think, it might be beneficial for students if the definitions of these terms, even some basic theories in complexity theory, can be introduced at the first class .

Bob, am I still allowed to ask you questions about simulation in future?


Hi Bob,

Great course and introduction to simulation! My comments:

  1. Personally I did not enrol because the assessment (I thought) was very intense and I could not guarantee completion in time. To encourage more enrolment perhaps assessment could be altered for next time, or eliminated all together with participation/presentations the only method of assessment (ie Pass/Fail)
  2. I think the course should be offered as an ECON course in Computational Economics. Perhaps it should be 1 session long so as to offer more chance to learn a little netlogo, and practice simulation

My goal was to think about possible Phd(econ) research topics utilising simulation techniques. I will be reviewing the online material and the texts over the summer to try and narrow the field further. To help narrow the search, I would appreciate your advice as to what are fruitful and interestng areas of research, and which could be practically completed over 2 years part-time at phd standard.

What interests me are:

  1. Trading systems:
    I wrote a trading system based on genetic-optimisation a few years ago, but never really tested it. I would be interested in identifying and applying the latest techniques like neural nets to do the same.
  2. Business Economics:
    I work in telecommunications, and I am trying to find some business data or behaviour that could be modelled via simulation. I have access to some business and marketing data (eg product statistics) that could be modelled and have some predictive power. But my imagination is lacking about exactly I could do.
  3. Economic Theory and Simulation.
    I thought The paper on risk aversion was a great example of how traditional economic theory is extended by simulation. Are there other examples like this were economic theory goes only part of the way to explaining micro or macro behaviour, but through simulation you see much more information about the behaviour of agents or societies.

Wondering if you can direct me to some books or articles which will allow me to explore the above areas further. Alternatively, if you have any topics in your mind, pleading for further research I would be interested to know.


Dear Bob,

I really enjoyed the class and was very disappointed that I couldn't make it to the last class. The evolutionary stuff would have been particularly interesting to discuss.

One suggestion I would possibly make is that there be more hands-on experience with building some models. My background as a software developer stood me in fairly good stead to understand some of the mechanics and limitations of simulation but, from some of my conversations with other students in the class, I don't think this was universal. I think getting your hands dirty really facilitates the development of some intuition about this sort of modeling. I don't know how appropriate that is for a PhD level course but I think it would have helped some of the participants achieve a deeper understanding of the topic.

I was also wondering whether you were still interested in meeting to discuss how I might use simulation to in my research. Sorry about my very confused presentation of my topic. It was partially due to the fact that I'm used to explaining my topic to people from Information Systems, who have very different backgrounds to yourself, so I wasn't quite sure where to begin. Anyway, if you have time that would be great.


Dear Bob,

Sorry to respond late. Can I add something as well?

I'm really grateful for being given this opportunity to learn from the ABM course delivered by you. For sure it opened a wonderful door for my current thinking for my research, which focuses on the efficiency, effectiveness, and resilience of parallel markets. In other words, I'm working on finding something out from the underlying reasons (personal motivation to get involved in the new patterns of purchasing behaviour) and the emergence of the new-formed market's impact to benefit different consumer groups or even the whole society. I'm not flattering you by saying that it's been so helpful and significant because there exists a huge gap between micro-marketing research (e.g., consumer behaviour and profit made by companies as decision-making units) and macro-marketing thinking (e.g., effectiveness to booster people's wellbeing and other overall social effects) in consistency of methodology for quite a long time. And I found myself facing a big challenge to establish a convincing way to bridge this gap between individual motivation and aggregate level until I knew more about the principles and applications of ABM in social science. For certain, every tool has its strengths and weak points. Controversial as it is, as a method, ABM is still believed to play a substantially critical role in the theoretical development as in my case. And I would integrate this powerful tool into my field.

The second thing I'd like to talk about is parsimony and the representativeness of ABMs. Researchers always hope to defend their methodology in the most convincing manner, through not only making the break-through in abstract theoretical meaning but also capturing those vivid features of the real world. I'd thought the combination of empirical data as wide and true as possible might reduce the risk of less-realism. However, as I was told in class, the involvement of superfluous information as initial settings, although in actual existence, will make the main effects dim! Actually it also needs a trade off between what is to be modelled (internal validity) and what is convincing (external validity).

As for the hotly debated presence of sufficiency and necessity in the course of scientific reasoning, I also face the problem of how to clearly define the boundary and then to exhaust the possibilities inside a system.

I haven't yet begun to expect details of the technical resolution as for the programming issues from this course since I deem this course fundamentally suitable for the beginners in ABMs and more meaningful in the understanding of this method. I may attach more expectation on the later summer school to be held at CSU to address these issues.

Presently, I'm still working on it, and hopefully approaching the truth as expected or with surprise. I'll continue definitely, probably still confused, but better.

Last but not least, address my personal respect to your work!

Best best regards,

Hi Bob,

Thanks a lot for a course that allowed me to "open up my thinking to new possibilities". Now that is a pretty big statement to make in science but it's one that I can certainly say is true.

I've realized that life and social situations can be more complex than they seem to from a more traditional perspective. Does this mean that I could come up with a brilliant idea when explaining this course's content to colleagues who have not been exposed to this kind of thinking? Could I come up with the magician's rabbit - and explain it all? Certainly not.

But I've learnt to be more curious, to be more willing to look at the relationships between "agents" (in psychology, we usually call them "people", but "agents" or - even better - "turtles" is fine with me too, haha) and whether these relationships might explain patterns on a higher level. So in a way, this course has given me the awareness that there is another perspective from which I should examine apparent mysteries from time to time...I've realized that subtle changes in one part of a system can bring about huge changes in other parts of the system.

Thanks again for a wonderful course! Throughout my university education I've usually regretted showing up for classes, feeling like I'm wasting my time. With this course it was the other way around, I couldn't attend the last two sessions and felt like I was missing out on something very important. Well, I think the bug has bitten me and hopefully will be able to apply some simulations in my own research in the future.