Wiki source code of BANOCOSS2023 Workshop

Last modified by Maria Valaste on 2024/04/16 14:30

Show last authors
1 == [[Home >>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/]]| [[Aim and scope>>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20AIMS%20AND%20SCOPE/]] | [[Call for Papers>>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Call%20for%20Papers/]] | [[Registration >>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Registration/]]| [[Important Dates>>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Important%20Dates/]] | [[Programme >>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Programme/]]| [[Workshop>>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Workshop/]] | [[Speakers >>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Speakers/]]| [[Instructions for Speakers>>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Instructions%20for%20Speakers/]] | [[Travel and Accommodation>>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Travel%20and%20Accommodation/]] | [[Sponsors >>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Sponsors/]]| [[Committees >>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Committees/]]| [[Contacts>>url:https://wiki.helsinki.fi/xwiki/bin/view/BNU/Events/BANOCOSS2023/BANOCOSS2023%20Contacts/]] ==
2
3 = Workshop on Multilevel regression and poststratification (MRP), 21 August 2023, 12 to 4 pm =
4
5
6 //The half-day workshop on Multilevel regression and poststratification (MRP) will take place as part of the conference on Monday 21 August. Participation (onsite and online) is free of charge.//
7
8 Speaker: **[[Philipp Christian Broniecki>>url:https://philippbroniecki.com/||shape="rect" class="external-link"]], University of Oslo, Norway**
9
10 autoMrP is a tool for estimating public opinion in subnational units, such as states, from nationally representative surveys. It combines multilevel regression with post-stratification (MrP) and machine learning. MrP has become the gold standard for small area estimation. Researchers who want to apply MrP are faced with three problems: how to select features, how to specify the functional form, and how to regularize the model parameters. autoMrP, an R package available on CRAN ([[https:~~/~~/cran.r-project.org/package=autoMrP>>url:https://cran.r-project.org/package=autoMrP||shape="rect" class="external-link"]]), addresses these problems in a systematic way using machine learning techniques. The method is described in detail in Broniecki, P., Leemann, L., & Wüest, R. (2022). Improved Multilevel Regression with Poststratification through Machine Learning (autoMrP). //The Journal of Politics//, 84(1), 597-601.
11
12 Time: Monday 21 August 2023, 12 to 4 pm ([[Time zone: UTC+3>>url:https://time.is/UTC+3||shape="rect" class="external-link"]])
13
14 Place: [[Soc&kom, Festsal>>url:https://tilavaraus.helsinki.fi/en/city-centre/swedish-school-social-science-snellmaninkatu-12/sockom-festsal||shape="rect"]], University of Helsinki and Zoom
15
16 Registration form (NOTE. If you have already registered for the BaNoCoSS conference, you do not need to register again.): [[https:~~/~~/elomake.helsinki.fi/lomakkeet/124015/lomake.html>>url:https://elomake.helsinki.fi/lomakkeet/124015/lomake.html||shape="rect"]]. Registration is required for both in-person and online (virtual) participants. Link to the online platform used to share materials etc. will be communicated and opened for registered participants before the event.
17
18 **Bring your own laptop. Welcome!**
19
20 **Dropbox link: [[https:~~/~~/bit.ly/3KOQbCa>>url:https://bit.ly/3KOQbCa||shape="rect"]]**
21
22 == **Outline autoMrP hands-on course** ==
23
24 Have [[R installed (preferably 4.x)>>url:https://www.r-project.org/||shape="rect"]]. We install packages from CRAN. In section 4 of the course, we apply autoMrP. Bring your questions and/or data if you have thought about working with autoMrP in your project.
25
26 1) MrP, regularization, and the problem autoMrP attempts to solve (12 – 12.50 pm)
27
28 * Review the MrP approach, the random intercepts model with post-stratification
29 * How partial pooling regularizes
30 * The problem that autoMrP attempts to solve
31 * Practical: In R, estimate an MrP model using the autoMrP package
32
33 10 minutes break
34
35 2) Machine learning for prediction tasks (1 – 1.50 pm)
36
37 * Review central concepts: bias-variance trade-off, cross-validation, loss-functions
38 * Review best subset, k nearest neighbors, the lasso, support vector machine, gradient boosting
39 * Review ensembles
40 * autoMrP recipe
41 * Practical: Use autoMrP
42
43 10 minutes break
44
45 3) autoMrP in practice I (2 – 2.50 pm)
46
47 * how to construct the census data
48 * how to input custom folds
49 * uncertainty
50
51 10 minutes break
52
53 4) autoMrP in practice II (3 – 4 pm)
54
55 * complex sampling designs (weighting, clustering)
56 * lopsided dependent variables
57 * when not to use autoMrP and how to tell
58 * alternatives to autoMrP
59 * questions/solving problems with your data
60
61
62
63 ~-~--
64
65 //Last update 21 August 2023//