search

DistanceDevelopment / spatial-workshops

Distance sampling workshop content
http://distancesampling.org/workshops/
2 stars 0 forks source link

Provisional workshop syllabus/schedule #42

Closed erex closed 7 years ago

erex commented 7 years ago

Created in January of 2017 image

dill commented 7 years ago

Alternate-ish timetable from @erex and me from last week (since it doesn't render well in syllabus.md on github):

screen shot 2017-06-19 at 17 31 12

erex commented 7 years ago

Practical materials to acquire/construct

LHMarshall commented 7 years ago

@lenthomas you said "we plan to generate a sample dataset to analyze on Day 1 using a simple dsm script, and to teach dsm (and hence a bit of automated survey design) on the afternoon of Day 1." Did you mean to say dsm script? Or DSsim? will this be done in Distance or R?

Things seem to have changed quite a bit? I might need some further explanation, what would you like me to teach and how does it fit in and link with other stuff?

I can't see how Eric's practicals fit with the timetable laid out by Dave for the Monday... there doesn't seem to be time to do the simulating data on Monday morning and maybe too much time on the simulation in the afternoon (but then you also mentioned that there would be survey design on Monday afternoon too?). The content of my simulation "lecture" is more a tour of the functionality. I assume the "sampling along trails" refers to the existing simulation exercise. I had thought that the simulation with covariates/group size would be the one used in the morning (rather than the afternoon) to generate the data fro analysis.

lenthomas commented 7 years ago

@LHMarshall Good points, my bad explanations. The plan is not too different from the original - I'll try to explain it a bit better.

Day 1 session 1 - I'll chair an interactive, student-led recap of the basics of CDS (basic formulae, finding p), assumptions, what to do when assumptions go wrong, possibly MCDS (to be determined).

Day 1 session 2 - students will work through datasets in Distance for Windows or Distance doing analyses. Datasets will start easy and end harder. For the easy dataset, one idea would be to generate a very straightforward simulated dataset, and then analyze it. The idea here is not to get bogged down into the minutae of DSsim (yes I meant DSsim, not dsm, above, sorry) -- but to get a dataset with known properties - first one perhaps with no covariate, tracklines given to them, etc. We'd provide a simple script (or workplan for Distance for windows) with some explanation, students run it, and voilla, some data. The purpose of this session in general is for students to re-acquaint themselves with analysis, and for us to communicate tips for analysis of gnarly data.

Day 1 session 3 and 4. This is where we do focus on DSsim. To do this, we need to first communicate automated survey design -- but the focus will be on doing this so as to generate simulated surveys rather than for the sake of automated survey design. I wasn't sure if you were able to/want to be involved in writing the contents of this. If yes, exactly how to play it is up to you. If no, then Eric was going to go for something based a fair bit on current materials. Up to you, and no pressure!

Hope this is clearer; happy to hear any feedback and thoughts. Also happy to meet face-to-face if you're able and want to (but again, no pressure)!

LHMarshall commented 7 years ago

@lenthomas ok might pop in to discuss survey design stuff with @erex at some point... Eric are you about tomorrow (Thursday)? Also I think it would be better for those involved in teaching the morning stuff to generate the project/script for data generation so it meets the requirements.

erex commented 7 years ago

@LHMarshall I'm here all day.

You noted in this issue that you will be unavailable from 07Jul until the workshop begins. Therefore the two weeks between now and 07Jul is the time to produce practicals and datasets for the "simple simulation" session 2 and "detailed" simulation session 3/4.

I see the survey design as subservient to performing simulations. To set up simulations, participants will need to touch the survey design engine. But I would rather the survey design business be a means to an end rather than and end to itself, but that is for you to decide.

As you say, you will be presenting the simulation practicals; happy for you to prepare them, or for you to tell me what to prepare. Either way, preparation ought to commence soon.

erex commented 7 years ago

A working version of the practicals for the workshop currently resides at https://rexstad.updog.co/dsm

Intent is to have all practicals integrated into this single HTML material, which can also be downloaded as PDF for the participants.

If possible, practicals in Rmd format can be bolted into this framework

erex commented 7 years ago

For reference, here are the rmarkdown exercises used during the 2016 workshop. Available to be recycled for 2017

There are also cds and mcds markdown exercises, but I don't believe there is room in the current Monday morning schedule for them.

erex commented 7 years ago

Research talks

dill commented 7 years ago

I think Joyce should give an inlabru talk, if she is willing/available.

I'd be happy to do the "other DSM approaches" stuff either way.

Since we have already some interest in designed experiment stuff, should we ask Steve/Cornelia to talk about that?

On 19/07/2017 10:56, erex wrote:

    Research talks
  • Glennie invited to deliver movement
  • Borchers to deliver SCR and HMM
  • Both accepted on 21Jun17
  • Nobody invited to deliver "other DSM approaches" o talk of this title customarily delivered by LJT during advanced workshop o is this to be replaced by inlabru talk?
    • then an invitation to Joyce should be sent

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/DistanceDevelopment/spatial-workshops/issues/42#issuecomment-316334765, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAHoeeloxkCExNKBmI3kFKySkKKCtQGks5sPdLQgaJpZM4N8PQN.

erex commented 7 years ago

I'll invite Joyce. We already have 5 talks slotted for 15 minutes, that invariably will take 20. I think nearly 2 hours of "peripheral" material is sufficient. Prefer not to have any more research talks. But my opinion is often in the minority.

lenthomas commented 7 years ago

Dave, I'm happy if you want to do other DSM approaches, or I can -- just let me know which you prefer. Cheers, Len

On 19/07/2017 12:05, DL Miller wrote:

I think Joyce should give an inlabru talk, if she is willing/available.

I'd be happy to do the "other DSM approaches" stuff either way.

Since we have already some interest in designed experiment stuff, should we ask Steve/Cornelia to talk about that?

On 19/07/2017 10:56, erex wrote:

Research talks

  • Glennie invited to deliver movement
  • Borchers to deliver SCR and HMM
  • Both accepted on 21Jun17
  • Nobody invited to deliver "other DSM approaches" o talk of this title customarily delivered by LJT during advanced workshop o is this to be replaced by inlabru talk?
  • then an invitation to Joyce should be sent

— You are receiving this because you commented. Reply to this email directly, view it on GitHub

https://github.com/DistanceDevelopment/spatial-workshops/issues/42#issuecomment-316334765,

or mute the thread

https://github.com/notifications/unsubscribe-auth/AAAHoeeloxkCExNKBmI3kFKySkKKCtQGks5sPdLQgaJpZM4N8PQN.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/DistanceDevelopment/spatial-workshops/issues/42#issuecomment-316337331, or mute the thread https://github.com/notifications/unsubscribe-auth/AAH57_ddxzSrkECDPD9TWN6wxaHLUQh8ks5sPdUFgaJpZM4N8PQN.Web Bug from https://github.com/notifications/beacon/AAH57zkSWyjb7w0q7i_zqDcdrd455GZOks5sPdUFgaJpZM4N8PQN.gif

{"api_version":"1.0","publisher":{"api_key":"05dde50f1d1a384dd78767c55493e4bb","name":"GitHub"},"entity":{"external_key":"github/DistanceDevelopment/spatial-workshops","title":"DistanceDevelopment/spatial-workshops","subtitle":"GitHub repository","main_image_url":"https://cloud.githubusercontent.com/assets/143418/17495839/a5054eac-5d88-11e6-95fc-7290892c7bb5.png","avatar_image_url":"https://cloud.githubusercontent.com/assets/143418/15842166/7c72db34-2c0b-11e6-9aed-b52498112777.png","action":{"name":"Open in GitHub","url":"https://github.com/DistanceDevelopment/spatial-workshops"}},"updates":{"snippets":[{"icon":"PERSON","message":"@dill in #42: I think Joyce should give an inlabru talk, if she is willing/available.\n\nI'd be happy to do the \"other DSM approaches\" stuff either way.\n\nSince we have already some interest in designed experiment stuff, should \nwe ask Steve/Cornelia to talk about that?\n\nOn 19/07/2017 10:56, erex wrote:\n\u003e \n\u003e Research talks\n\u003e \n\u003e Glennie invited to deliver movement\n\u003e Borchers to deliver SCR and HMM\n\u003e Both accepted on 21Jun17\n\u003e Nobody invited to deliver \"other DSM approaches\"\n\u003e o talk of this title customarily delivered by LJT during advanced\n\u003e workshop\n\u003e o is this to be replaced by inlabru talk?\n\u003e + then an invitation to Joyce should be sent\n\u003e \n\u003e —\n\u003e You are receiving this because you commented.\n\u003e Reply to this email directly, view it on GitHub \n\u003e \u003chttps://github.com/DistanceDevelopment/spatial-workshops/issues/42#issuecomment-316334765\u003e, \n\u003e or mute the thread \n\u003e \u003chttps://github.com/notifications/unsubscribe-auth/AAAHoeeloxkCExNKBmI3kFKySkKKCtQGks5sPdLQgaJpZM4N8PQN\u003e.\n\u003e \n"}],"action":{"name":"View Issue","url":"https://github.com/DistanceDevelopment/spatial-workshops/issues/42#issuecomment-316337331"}}}

-- Len Thomas len.thomas@st-andrews.ac.uk lenthomas.org @len_thom Centre for Research into Ecological and Environmental Modelling The Observatory, University of St Andrews, Scotland KY16 9LZ Office: UK+1334-461801 Admin: UK+1334-461842

The University of St Andrews is a charity registered in Scotland, No SC013532.

dill commented 7 years ago

Live version of schedule here:

http://distancesampling.org/workshops/stand-intermed-2017/

Closing this issue.