Geoff Coyle

Geoff Coyle passed away November 19.  As Tom Fiddaman described on his blog, Geoff was a pioneer in system dynamics and certainly seemed to qualify for the term irrascible, as Tom noted, for he had his own views and a clear way of stating them.

He was very willing to share his insights.  He and I met on the old system dynamics mailing list. Through private emails, we discovered that we shared the belief that the engineering approach calls for varying levels of models, ranging from rules of thumb through the simple, highly-aggregated qualitative models all the way, in the appropriate cases, to carefully done simulation models.  It's not the case that every problem demands detailed simulations, nor is it the case that a simple causal loop diagram (or influence diagram, as he preferred) always suffices.  It was the same when I was an analog circuit designer: sometimes I would design a circuit based on simple rules of thumb, while other circuits called out for detailed analysis, and that had been his experience in his world, as well.

I learned a lot from Geoff: the rigorous use of influence diagrams, the use of QPID, the Analytic Hierarchy Process for decision making, and improved modeling skills at all levels of rigor.  I read his Management Systems Dynamics, and I read and taught from his System Dynamics Modelling: A Practical Approach.

In recent years, we wrote less, for he was more focused on his latest book on mining, but I'm glad to have known him.  I invite anyone who is interested in system dynamics to check out Tom's list of Geoff's writings. I certainly learned from them; perhaps you will, too.

Thank you, Geoff, for sharing your insights and engaging in dialog.

More Exploratory Data Analysis

I missed two items in my last posting on EDA.  First, there's another tool I use: J.  Over time, I seem to oscillate between R and J.  J, at least when I don't let myself get rusty in expressing ideas in J, is a powerful and concise way of thinking.  R has an enormous library collection.  I won't advocate for one over the other here, but you should try both.

Second, some tend to think of EDA as model-free statistics, but that's not quite right.  To get a better explanation of what that means, see Andrew Gelman's A Bayesian Formulation of Exploratory Data Analysis and Goodness-of-fit Testing.I'd go past what he wrote and note that using models in EDA extends to work in system dynamics.  To make sense of a dynamic (time-varying) situation, often trying to craft a model that approximates the situation is a great way to get started in making sense of the situation.

Exploratory Data Analysis

I've used exploratory data analysis (EDA) in my work and in my thinking about situations ever since discovering it years ago.  I've used paper and pencil, XLISP-STAT, R, ggobi, and other tools.  I've read a number of the books.

When suggesting EDA to others, I've been puzzled: do I recommend they read one of the books, which is likely more than they want to start?  Do I tell them about particular techniques, which miss the flavor and contribution of EDA?

A year or two ago, I discovered John Behrens' Principles and Procedures of Exploratory Data Analysis, which seems to contain principles and some of the common procedures. I'm noting this here, both for you to find and for me to re-find more easily in the future.

In searching for this article, I ran across Chong-ho Yu's site on Exploratory data analysis and Data visualization, which seems to provide a good introduction to the ideas of EDA as expressed by Behrens.  That doesn't seem surprising, as Yu seems to have studied under Behrens at ASU.

One Economist's View of Growth

Northwestern University's Robert Gordon just published Is US economic growth over? Faltering innovation confronts the six headwinds, a provocative article I recommend reading.  In it, he shows data that supports declining US economic growth since the 1950s, and he provides a framework for understanding it.

What he writes seems related to our earlier discussions on growth and to The Club of Rome's work on the limits to growth 40 years ago.  While that work pointed to an impending clash between exponential growth and finite resources, Gordon's points to evidence of slowing growth in the economic data.  His graph of UK/US economic growth from 1300 to 2007 (with predictions out further) reminds me of Hubbert's pimple on the use of petroleum as a fuel.

There's a difference between Gordon's thesis and that of the Club of Rome, though.  Gordon's article shows S-shaped growth that is coming to an end, while the Club of Rome work warns of potential overshoot.  

Thanks to Mark Thoma for the pointer.

Credibility in modeling

System dynamicists have multiple approaches to testing models; you can read chapter 21 of John Sterman's Business Dynamics for a good introduction.

Or you can read Paul Krugman's Who To Listen To. I like his attitude, and it sounds remarkably like a system dynamicist wrote it.

It's true, of course, in statistics, too.  Andrew Gelman has written about that multiple times, and he alludes to the importance of the model and its behavior in Ripley on model selection, and some links on exploratory model analysis.

What do practitioners need to know about regression?

That's a question I sometimes get asked.  Apparently Andrew Gelman does, too, for he posted What do practitioners need to know about regression? in 2010.  I think it's still valuable, if a bit cryptic.  He does reference the book he co-authored with Jennifer Hill, and that should answer most of anyone's remaining answers, at least at this level.

Plot the data!

I've heard the injunction "Plot the data" for decades.  Plotting the data is important throughout statistical analyses, from initial looks to surmise what model might fit the data to looking at residual plots to check the model or see what might be a good next step. Plotting the reference mode or reference behavior pattern of a problem is a key beginning step in the application of system dynamics.

At first, graphs were done manually with colored pencils and graph paper, then there were HP's great pen plotters, and then we got today's almost universal system capability of displaying and printing high-resolution graphics.  We're indebted to those such as William Cleveland and Edward Tufte for their research-based insights on ways to generate helpful, honest graphics.

I also understand the utility of tables of data, at least in the right circumstances. In that regard, I like Andrew Ehrenberg's advice in his The Problem of Numeracy, and his advice on reading tables is worth reading, too.

Now there's a bit of research that may help in the decision whether to use a graph or a table for displaying data.  Economists Are Overconfident. So Are You from the HBR Blog Network points to research that indicates that economists do better at understanding regression analyses when only shown graphs than when shown graphs and numbers and much better than when only shown numbers.

So plot the data already!

Making sense of lots of data

I'm a fan of William Cleveland's insights into what works in graphics.  His ideas have spread widely; you may have seen some in Tufte's work.

Recently I ran across a relatively new way of displaying time series graphs.  It's called a horizon plot.  At first, I thought it rather contrived.  As Stephen Few also decided, they are beginning to grow on me.  Read his Time on the Horizon for a good introduction or Timely Portfolio's Horizon Plot Already Available for how to create them in R. Panopticon, their developer, has Visualize Multiple Data Sets & Compare Trends with Horizon Graphs, Thanks to the University of California, Berkeley, for their Sizing the Horizon: The Effects of Chart Size and Layering on the Graphical Perception of Time Series Visualizations, which led me to find them.


I think you may see more of these in the future, so spend a "Few" minutes (pun intended) to learn how to read them.  

Improving your chances

What can we each do to improve our chances of making a professional contribution over time? What keeps us employable? There are no guarantees, but AnnMaria De Mars gave her views recently in Why I Won’t Be Unemployed in 2017: It’s Apples. The punch line comes at the end, but you need to read from the beginning to understand the message.

What do you think? What's your approach to learning, now that (most of you, probably) are out of formal schooling?

Changes

The main site for Facilitated Systems will be undergoing various changes over the next period of time, as I restructure it and modify the content somewhat. There's one important change to note: in the past, you could access any of the Facilitated Systems pages except this blog with http://facilitatedsystems.com. Starting soon, you'll need the full name: http://www.facilitatedsystems.com/. If you try the former and it doesn't work, try the latter. If you have bookmarked a page somewhere inside my site and it fails to work, try the home page, and work from there.

A Story of Three Books

I just read two books, which made me think of a third to add context.

I first read Management der Unternehmensentwicklung: Phasengerechte Fuhrung und der Umgang mit Krisen (St. Galler Management-Konzept) (German Edition) when the book was first published. I found it a helpful book in that it explained nicely the lifecycle of a firm and how one should act effectively in each of the four phases they described. One thing jumped out at me: as a firm approaches the last phase, it needs to think about returning to the first. That's a problem, because startup phases won't support the full organization, so the typical response is to split the firm into different segments or business units. These two (or more) units take different times to pass through the phases, so that the firm eventually finds itself with different parts in different phases and thus needing different approaches to management. That flexibility is difficult to manage. I recommend this to any who are interested in management of organizations and who can read German.

Much more recently, I finally got around to reading The HP Phenomenon: Innovation and Business Transformation (Stanford Business Books). Many management histories I've read tend to praise or condemn the past, but House and Price did a very good job of portraying the challenges of the major transformations Hewlett-Packard has made since it was founded in 1939. I learned much about what I had not observed while working there; I also learned to think more deeply about the challenges we all see in organizations. I'd recommend this to any who have been or are still at HP, for you may gain perspective on your experiences. I would also recommend it to any from other high-tech, growth-oriented companies, for it might help you think about what you'll likely face as your company evolves.

Finally, I just finished David Wortley's Gadgets to God. Like The HP Phenomenon, this book focuses on transformations, this time at a much smaller scale. As Chuck and Ray did for large companies, David gives a clear description of the challenges and transformations he negotiated as an entrepreneur of a small business in a growing, changing market. As someone who ran a small business in much the same time, I find that his descriptions of the life of a small businessperson should be required reading for anyone setting out to do the same thing, for it should help you prepare for the challenges you'll face. David also puts the developments of the past few decades in a larger context, which may help you, too, consider the values and norms you bring to your work. Incidentally, David is a friend and colleague, for, as you can find somewhere in the book, we've worked together for a number of years without ever having seen each other.

Three books, one theme: transformations. The theme is a seeming constant in our world, and each of the three might bring different facets of the theme to light in your work.

Running into a brick wall

When I teach information dynamics at the University of Washington (what their Information School calls system dynamics), I include material on initializing models in equilibrium. That is a common technique for creating models that are easier to understand and easier to make work as intended.

I go past the current text, Sterman's Business Dynamics: Systems Thinking and Modeling for a Complex World with CD-ROM
, to include techniques from Richardson's and Pugh's Introduction to System Dynamics Modeling with Dynamo. They teach how to initialize a model in a state of constant growth, too. That's handy for certain types of models, and it teaches a useful principle.

To sketch out their example, imagine a population of anything that can be modeled with a simple birth and death process. Births per year is proportional to current population, and deaths per year are modeled as the population divided by the average lifespan. You can work out the equations or read chapter 4 of Richardson and Pugh to figure out how to initialize a model in a state of constant growth.

Imagine you've done that for an arbitrary growth rate. If there is a limit to growth, then, at some time, births per year will exactly equal deaths per year. At that point, the net growth is zero. If you think of world population and birth rates of 4% to 5% per year (the figures quoted in the book), and if you assume that the limit to growth occurs because of an increase in deaths per year, a bit of algebra lets you compute the average lifespan at that moment of equilibrium as 20-25 years. When students discover this, the classroom usually gets very quiet as the lesson sinks in.

As the lesson comes from the equations and their behavior, it would apply to any system that has birth and death processes, whether it's a population of living things or an economic system such as our economy of businesses.

That lesson is important and the idea, while from a simplified model, is solid. Still, it's nice to see whether a similar situation occurs in the real world. Miller-McCune just published "Foreclosure Forcefield" in their January/February 2012 issue, ponting to research by Garrett Glasgow at the University of California, Santa Barbara, indicating that cities with "slow-growth policies in place to prevent rampant residential construction" fared better in the foreclosure crisis than did those that were "aggressively pro-growth." That sounds a lot like the lesson from Richardson's and Pugh's example. You can see more in "Local Development Policies and the Foreclosure Crisis in California: Can Local Policies Hold Back National Tides?" by Glasgow, Lewis, and Neiman in Urban Affairs Review, 48(1), pp. 62-83.

Or, in simple English, if you think you might hit a brick wall, it's much better to be walking slowly than running as fast as you can.

Learning logs

For some years, the learning logs Bob Williams and I assembled have been one of the popular pages on this Web site. Whether you use that format or another, I continue to see great value in reflection as an action learning tool, and I see value in journaling in supporting reflection.

Still, the reason I'm posting this today is that I found an old Dilbert cartoon in my learning log notes that I thought you might enjoy.

Instead of making a New Year's resolution, consider how you'll learn from your experiences, and consider reflection and journaling as part of your approach.

Thinking for yourself

Some years ago, I took freshman chemistry with Dr. Zevi Salzberg at Rice University.  I still recall one lecture that made a fundamental impression on me about how one learns and works as a professional.

As I recall, he was going to teach us about solubility and pH calculations.  He stepped to the front of the class and said something like, "Well, I forgot my notes today, so let's see if I can re-derive everything I was going to present here instead of taking time to go back to my office to get them."  Then he spent the rest of the class working through the material from (close to) first principles.   By the end of the class, I understood how to do those calculations, and I remembered that for a long time, even though I never took another chemistry course after my freshman year.

More importantly, though, I understood that one can and often should work through things for oneself. For one reason, it was possible.

For another, better reason, one was arguably more able to understand a limited set of first principles in a field and then work from that understanding to solve a problem than to remember all the specialized formulas and procedures to solve any of a large set of types of problems. Besides, remembering all those formulas and procedures wouldn't help much when faced with a new problem, but remembering first principles would.

For a final reason, it would help one continue to learn in whatever situation one faced in one's career.

I was reminded of that yesterday when I read Brad DeLong's reposting of Joan Robinson's "Open letter from a Keynesian to a Marxist," as given in Mike Beggs' piece on the Jacobin blog.

I'm not here today to persuade anyone to become a Keynesian or a Marxist; I'm writing because Robinson sounds like she has the same approach as Dr. Salzberg, when she writes,

The thing I am going to say that will make you too numb or too hot (according to temperament) to understand the rest of my letter is this: I understand Marx far and away better than you do. ...

When I say I understand Marx better than you, I don’t mean to say that I know the text better than you do. If you start throwing quotations at me you will have me baffled in no time. In fact, I refuse to play before you begin.

What I mean is that I have Marx in my bones and you have him in your mouth.

Read her letter, and see what you think.  Does she make sense to you about her way of knowing? She does to me, as did Dr. Salzberg.

By the way, I have no good reason to believe that Dr. Salzberg really forgot his notes that day; I fully surmise it was a pedagogical ruse to teach us exactly that message, and it worked. If he did really forget, I hope he forgot other years, too.

I noted one other paragraph in her letter: "But I want you to think about me dialectically. The first principle of the dialectic is that the meaning of a proposition depends on what it denies. Thus the very same proposition has two opposite meanings according to whether you come at it from above or from below. I know roughly from what angle you come to Keynes, and I quite see your point of view. Just use a little dialectic, and try to see mine." I think we'd be well served to think about that when we face some of the contentious issues of our time. She didn't ask him to adopt her views; she simply asked him to try to understand them, if for no other reason that it would help him understand his views, in some ways the antithesis of hers, better.