Pointers to papers/books on DOE & ABM?
Locked
Pointers to papers/books on DOE & ABM?
 
Administrator
|
Anyone have good pointers to papers or books on Design of
Experiments, mainly with an eye towards choosing good control parameters for Agent Based Modeling? -- Owen Owen Densmore http://backspaces.net - http://redfish.com - http://friam.org |
Pointers to papers/books on DOE & ABM?
|
|
I don't know if it has exactly what you are looking for, but I think
Nigel Gilbert's text on Simulation for the Social Scientist, 2nd Edition, is filled with good resources on modeling. There are also a slew of standard research methods texts out there with stuff on research design, but I would also like to see references specific to agent-based modeling. Mike On 3/11/06, Owen Densmore <owen at backspaces.net> wrote: > Anyone have good pointers to papers or books on Design of > Experiments, mainly with an eye towards choosing good control > parameters for Agent Based Modeling? > > -- Owen > > Owen Densmore > http://backspaces.net - http://redfish.com - http://friam.org > > > > ============================================================ > FRIAM Applied Complexity Group listserv > Meets Fridays 9a-11:30 at cafe at St. John's College > lectures, archives, unsubscribe, maps at http://www.friam.org > |
Pointers to papers/books on DOE & ABM?
|
|
Owen:
I think I have a copy of Gilbert fairly close to the top of the stack on my desk. Let me know if you would like to take a look. I can bring it to Wedtech when, in fact, the very same Michael Gizzi is giving the presentation. -tj On 3/11/06, Michael Gizzi <mgizzi at gmail.com> wrote: > > I don't know if it has exactly what you are looking for, but I think > Nigel Gilbert's text on Simulation for the Social Scientist, 2nd > Edition, is filled with good resources on modeling. > > There are also a slew of standard research methods texts out there > with stuff on research design, but I would also like to see references > specific to agent-based modeling. > > Mike > > On 3/11/06, Owen Densmore <owen at backspaces.net> wrote: > > Anyone have good pointers to papers or books on Design of > > Experiments, mainly with an eye towards choosing good control > > parameters for Agent Based Modeling? > > > > -- Owen > > > > Owen Densmore > > http://backspaces.net - http://redfish.com - http://friam.org > > > > > > > > ============================================================ > > FRIAM Applied Complexity Group listserv > > Meets Fridays 9a-11:30 at cafe at St. John's College > > lectures, archives, unsubscribe, maps at http://www.friam.org > > > > ============================================================ > FRIAM Applied Complexity Group listserv > Meets Fridays 9a-11:30 at cafe at St. John's College > lectures, archives, unsubscribe, maps at http://www.friam.org > -- ============================================== J. T. Johnson Institute for Analytic Journalism www.analyticjournalism.com 505.577.6482(c) 505.473.9646(h) http://www.jtjohnson.com tom at jtjohnson.com "You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete." -- Buckminster Fuller ============================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20060311/8caab59d/attachment.htm |
Pointers to papers/books on DOE & ABM?
|
|
Here is a pdf of Simulation for the Social Scientist (c) 1999 version (274 pages)
http://perso.ens-lyon.fr/eric.boix/Cat/SimulationSurvey/Articles/GilbertTroitzsch1999SimulationForTheSocialScientist.pdf Lou The Meaning of Life ... Last words from M. Python.... Well, that's the end of the film. Now, here's the meaning of life. Thank you, Brigitte. M-hmm. Well, it's nothing very special. Uh, try and be nice to people, avoid eating fat, read a good book every now and then, get some walking in, and try and live together in peace and harmony with people of all creeds and nations, and, finally, here are some completely gratuitous pictures of penises to annoy the censors and to hopefully spark some sort of controversy, which, it seems, is the only way, these days, to get the jaded, video-sated public off their fucking arses and back in the sodding cinema. Family entertainment bollocks. What they want is filth: people doing things to each other with chainsaws during tupperware parties, babysitters being stabbed with knitting needles by gay presidential candidates, vigilante groups strangling chickens, armed bands of theatre critics exterminating mutant goats-- Where's the fun in pictures? Oh, well, there we are. Here's the theme music. Goodnight. ----- Original Message ----- From: Tom Johnson To: The Friday Morning Applied Complexity Coffee Group Sent: Saturday, March 11, 2006 6:26 PM Subject: Re: [FRIAM] Pointers to papers/books on DOE & ABM? Owen: I think I have a copy of Gilbert fairly close to the top of the stack on my desk. Let me know if you would like to take a look. I can bring it to Wedtech when, in fact, the very same Michael Gizzi is giving the presentation. -tj On 3/11/06, Michael Gizzi <mgizzi at gmail.com> wrote: I don't know if it has exactly what you are looking for, but I think Nigel Gilbert's text on Simulation for the Social Scientist, 2nd Edition, is filled with good resources on modeling. There are also a slew of standard research methods texts out there with stuff on research design, but I would also like to see references specific to agent-based modeling. Mike On 3/11/06, Owen Densmore <owen at backspaces.net > wrote: > Anyone have good pointers to papers or books on Design of > Experiments, mainly with an eye towards choosing good control > parameters for Agent Based Modeling? > > -- Owen > > Owen Densmore > http://backspaces.net - http://redfish.com - http://friam.org > > > > ============================================================ > FRIAM Applied Complexity Group listserv > Meets Fridays 9a-11:30 at cafe at St. John's College > lectures, archives, unsubscribe, maps at http://www.friam.org > ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org -- ============================================== J. T. Johnson Institute for Analytic Journalism www.analyticjournalism.com 505.577.6482(c) 505.473.9646(h) http://www.jtjohnson.com tom at jtjohnson.com "You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete." -- Buckminster Fuller ============================================== ------------------------------------------------------------------------------ ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20060311/60fd86fa/attachment-0001.htm |
Pointers to papers/books on DOE & ABM?
|
|
In reply to this post by Tom Johnson
Owen,
I also have a copy of Nigel's book.Hmmm, and the inside cover has your name on it. I'll return it on Monday :-) While it certainly is a good introduction to the social simulation space, a quick scan didn't reveal anything on Design of Experiments (DOE). Though, here is a paper also by Nigel Gilbert and Klaus Troitzsch (http://web.econ.unito.it/terna/deposito/gil_ter.pdf) where they do mention DOE in the context of ABM: "Thirdly, we need to evaluate whether the simulation is a success. A mathematical statistics approach can be used in which, according to Kleijenen (1998): The type of statistical test actually applied depends on the availability of data on the real system [that we are simulating]: (i) no data, (ii) only output data, and (iii) both input and output data. In case (i), the system analyst can still experiment 16 with the simulation model to obtain simulated data. Those experiments should be guided by the statistical theory on design of experiments (DOE); an inferior - but popular - approach is to change only one factor at a time. In case (ii), real and simulated output data may be compared through the well-known Student t statistic. In case (iii), trace-driven simulation becomes possible. This type of approach is potentially relevant, but often difficult to apply to an agent based computational model, where qualitative results are expected as well as quantitative ones. We therefore have to adopt a weaker criterion. One such is to compare the actions of the agent behaving in the simulated framework with our knowledge about the way that actual agents behave. A more severe degree of assessment is to test the aggregate effects of agents' behaviour in terms of emerging structure, groupings, spatial effects, and so on." Some other links on DOE in the context of simulation: * a presentation PDF by Nancy Gould : http://www.dtic.mil/ndia/2001systems/gould.pdf * an old IBM paper: http://tinyurl.com/e5aud * Robert Holmes wrote up the DOE bit in our DrugTalk paper with Mike: http://www.redfish.com/research/agent2004DrugSim.pdf * A more general introduction to DOE: http://www.coolingzone.com/Guest/News/NL_MAR_2004/Garron/Garron_Mar_04.html * a page of links on DOE: http://www.statease.com/articles.html Of the DOE books at Amazon (http://tinyurl.com/eq9wn), the 2004 Montgomery looks good. There's a used one for $69 that I would split with you. - S -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20060311/d99e3ea6/attachment.htm |
Pointers to papers/books on DOE & ABM?
|
|
In reply to this post by Tom Johnson
Personally, I wouldn't trust any entity whose initials were DOE.
DOH! Sorry, personal prejudices showing here. --Doug On 3/11/06, Stephen Guerin <stephen.guerin at redfish.com> wrote: > > Owen, > > > > I also have a copy of Nigel's book?Hmmm, and the inside cover has your > name on it. I'll return it on Monday :-) > > > > While it certainly is a good introduction to the social simulation space, > a quick scan didn't reveal anything on Design of Experiments (DOE). > > > > Though, here is a paper also by Nigel Gilbert and Klaus Troitzsch ( > http://web.econ.unito.it/terna/deposito/gil_ter.pdf) where they do mention > DOE in the context of ABM: > > > > "Thirdly, we need to evaluate whether the simulation is a success. A > > mathematical statistics approach can be used in which, according to > Kleijenen > > (1998): > > The type of statistical test actually applied depends on the availability > of data on > > the real system [that we are simulating]: (i) no data, (ii) only output > data, and (iii) > > both input and output data. In case (i), the system analyst can still > experiment > > 16 > > with the simulation model to obtain simulated data. Those experiments > should > > be guided by the statistical theory on design of experiments (DOE); an > inferior - > > but popular - approach is to change only one factor at a time. In case > (ii), real > > and simulated output data may be compared through the well-known Student t > > statistic. In case (iii), trace-driven simulation becomes possible. > > This type of approach is potentially relevant, but often difficult to > apply to an > > agent based computational model, where qualitative results are expected as > well > > as quantitative ones. > > We therefore have to adopt a weaker criterion. One such is to compare the > > actions of the agent behaving in the simulated framework with our > knowledge > > about the way that actual agents behave. A more severe degree of > assessment is > > to test the aggregate effects of agents' behaviour in terms of emerging > structure, > > groupings, spatial effects, and so on." > > > > > > Some other links on DOE in the context of simulation: > > > > - a presentation PDF by Nancy Gould : > http://www.dtic.mil/ndia/2001systems/gould.pdf > > > > - an old IBM paper: http://tinyurl.com/e5aud > > > > - Robert Holmes wrote up the DOE bit in our DrugTalk paper with > Mike: http://www.redfish.com/research/agent2004DrugSim.pdf > > > > - A more general introduction to DOE: > http://www.coolingzone.com/Guest/News/NL_MAR_2004/Garron/Garron_Mar_04.html > > > > - a page of links on DOE: http://www.statease.com/articles.html > > > > Of the DOE books at Amazon (http://tinyurl.com/eq9wn), the 2004 Montgomery > looks good. There's a used one for $69 that I would split with you? > > > > - S > > ============================================================ > FRIAM Applied Complexity Group listserv > Meets Fridays 9a-11:30 at cafe at St. John's College > lectures, archives, unsubscribe, maps at http://www.friam.org > > -- Doug Roberts, RTI 505-455-7333 - Office 505-670-8195 - Cell -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20060311/9ac4a4c0/attachment-0001.htm |
«
Return to Friam
|
1 view|%1 views
| Free forum by Nabble | Edit this page |