UseR Oslo
UseR Oslo
  • 21
  • 135 556
Real-Time Public Health Surveillance with Sykdomspulsen
Meetup link to the event: www.meetup.com/oslo-user-group/events/285821922/
[Abstract]
Sykdomspulsen is a real-time analysis and disease surveillance system designed at developed at the Norwegian Institute of Public Health (FHI). Sykdomspulsen processes new data collected from 15 data sources (e.g., covid-19 cases), runs 1000.000+ statistical analysis automatically for all locations (nation, county, municipality) in Norway, produces 1000+ reports and alerts for public health authorities and shares data to the public on GitHub.
Sykdomspulsen runs on a collection of R packages, the {splverse}. {splverse} is an ecosystem for infectious disease surveillance, from analysis planning, statistical analysis to reporting via visualization, shiny website and Rmarkdown generated reports. In this talk, Chi will present how Sykdomspulsen does public health real-time surveillance during the pandemic using R. Chi will introduce some of the core packages and illustrate how they work together, with an example using real surveillance data published daily on GitHub.
[Bio]
Chi is currently working at the Sykdomspulsen team as a researcher and R developer, at the Norwegian Institute of Public Health. Before she joined Sykdomspulsen in the middle of the pandemic (2020), she was a PhD student at the Department of Biostatistics at University of Oslo (OCBE), working on hospital EHR data.
Переглядів: 345

Відео

Missing Data Treatment in R: A Hands-on Illustration Using {mice}Missing Data Treatment in R: A Hands-on Illustration Using {mice}
Missing Data Treatment in R: A Hands-on Illustration Using {mice}
Переглядів 2,2 тис.2 роки тому
Join us for our first in-person event in over two years! www.meetup.com/Oslo-useR-Group/events/285126745/ If you want to code along to the talk, you will need a few packages. The can be installed using the following lines of code: # Install necessary package install.packages( c("mice", "VIM“, “MASS”, “lattice”), dependencies = T ) # Load the mice package suppressWarnings(suppressMessages( libra...
Bayesian Item Response Modeling in R with {brms} and StanBayesian Item Response Modeling in R with {brms} and Stan
Bayesian Item Response Modeling in R with {brms} and Stan
Переглядів 2,6 тис.2 роки тому
Recording from UseR Oslo's meetup April 7, 2022 - www.meetup.com/Oslo-useR-Group/events/284351144/ Item Response Theory (IRT) is widely applied in psychology and the social sciences to model persons' responses on a set of items measuring one or more latent constructs. While several software packages have been developed that implement IRT models, they tend to be restricted to respective prespeci...
Bayesian Latent Variable Modeling in R with {blavaan}Bayesian Latent Variable Modeling in R with {blavaan}
Bayesian Latent Variable Modeling in R with {blavaan}
Переглядів 3,2 тис.2 роки тому
Recording from UseR Oslo's meetup March 10, 2022 - www.meetup.com/Oslo-useR-Group/events/283674411/ The R package {blavaan} is an interface between package {lavaan} and MCMC software (JAGS and Stan), allowing users to specify a {lavaan} model that is then automatically estimated via JAGS or Stan. In the presentation, Ed provides an overview of {blavaan} and its use in practice. He shows some in...
Cluster Robust Standard Errors in R with {clubSandwich}Cluster Robust Standard Errors in R with {clubSandwich}
Cluster Robust Standard Errors in R with {clubSandwich}
Переглядів 2,4 тис.2 роки тому
Recording from UseR Oslo's meetup on February 3rd, 2022 - www.meetup.com/Oslo-useR-Group/events/283050203/ The slides used in this talk can be found here: www.jepusto.com/files/clubSandwich-Oslo-RUG-2022-02-03.pdf [About] Cluster-robust variance estimation methods (also known as sandwich estimators, linearization estimators, or simply "clustered" standard errors) are a standard inferential tool...
Wrapping Packages in R with {devtools} and FriendsWrapping Packages in R with {devtools} and Friends
Wrapping Packages in R with {devtools} and Friends
Переглядів 5682 роки тому
For this year's final meetup,we wrappeda holiday season-themed package! Using {devtools} and all its little helpers, like {usethis} and {testthat}, Raoul gave a walk through of the basic process of making a package, and addressed some of the most common best practices and dos-and-donts. The speaker, Raoul Wolf, is the main organizer of UseR Oslo, and a senior environmental advisor and scientifi...
Meta Analytic Structural Equational Modeling with {metaSEM}Meta Analytic Structural Equational Modeling with {metaSEM}
Meta Analytic Structural Equational Modeling with {metaSEM}
Переглядів 3,4 тис.2 роки тому
Recording from UseR Oslo's meetup on September 2nd, 2021 - www.meetup.com/Oslo-useR-Group/events/280005225/ Abstract: We often formulate models to understand how our data is connected. However, it is difficult to assess whether our model is a good representation of the data or the generalizability of our model to other contexts. In this lighting talk, we will explore meta-analytic structural eq...
Meta-Analysis of Nonparametric Models with {metagam}Meta-Analysis of Nonparametric Models with {metagam}
Meta-Analysis of Nonparametric Models with {metagam}
Переглядів 1,2 тис.2 роки тому
Recording from UseR Oslo's meetup on September 2nd, 2021 - www.meetup.com/Oslo-useR-Group/events/280005225/ Abstract: "Analyzing biomedical data from multiple studies has great potential in terms of increasing statistical power, enabling detection of associations of smaller magnitude than would be possible analyzing each study separately. Restrictions due to privacy or proprietary data as well ...
Calculating the Statistical Power of Studies Included in a Meta-Analysis Using {metameta}Calculating the Statistical Power of Studies Included in a Meta-Analysis Using {metameta}
Calculating the Statistical Power of Studies Included in a Meta-Analysis Using {metameta}
Переглядів 1,3 тис.2 роки тому
Recording from UseR Oslo's meetup on September 2nd, 2021 - www.meetup.com/Oslo-useR-Group/events/280005225/ Several methods exist for assessing study quality in meta-analysis, but these are typically discipline specific. One approach used across disciplines to assess the evidential value of an individual study is to calculate its statistical power, which can identify the range of effect sizes t...
Meta analysis of dependent effect sizes Robust variance estimation with {clubSandwich}Meta analysis of dependent effect sizes Robust variance estimation with {clubSandwich}
Meta analysis of dependent effect sizes Robust variance estimation with {clubSandwich}
Переглядів 2,9 тис.2 роки тому
Recording from UseR Oslo's meetup on September 2nd, 2021 - www.meetup.com/Oslo-useR-Group/events/280005225/ Abstract: Across scientific fields, large meta-analyses often involve dependent effect size estimates. Robust variance estimation (RVE) methods provide a way to include all dependent effect sizes in a single meta-analysis model, even when the nature of the dependence is unknown. RVE uses ...
Meta-Analysis in R with {metafor}Meta-Analysis in R with {metafor}
Meta-Analysis in R with {metafor}
Переглядів 40 тис.2 роки тому
The recording from UseR Oslo's meetup on August 26th, 2021 - www.meetup.com/Oslo-useR-Group/events/280005208/ Code used in the presentation can be found here: Overview of resources from Wolfgang Viechtbauer: www.wvbauer.com/doku.php/presentations Direct link to the code: www.wvbauer.com/lib/exe/fetch.php/talks:2021_viechtbauer_oslo_user_metafor.r Direct link to the slides: www.wvbauer.com/lib/e...
Build Interactive {shiny} Apps to Share Your Work With Anyone!Build Interactive {shiny} Apps to Share Your Work With Anyone!
Build Interactive {shiny} Apps to Share Your Work With Anyone!
Переглядів 1,4 тис.3 роки тому
This is the recording from the Oslo Use R meetup 20.05.2021 www.meetup.com/Oslo-useR-Group/events/277702734/ [Abstract] {shiny} is an R package that makes it easy to build interactive web apps straight from R. This is a powerful way to share your work in R with people without any prior knowledge. We will show how you can create and put such applications into production within a modern cloud inf...
Get Your Results Faster With {future}Get Your Results Faster With {future}
Get Your Results Faster With {future}
Переглядів 4973 роки тому
The recording from UseR Oslo's meetup on April 15th, 2021 - www.meetup.com/Oslo-useR-Group/events/277150580/ The {future} package offers a unifying framework to distributed computing in R. It lets you run exactly the same code - and achieve exactly the same results - sequentially on your laptop, in parallel on your laptop, in the cloud, or on large computing clusters. In this talk, Øystein will...
{ggplot2} Wizardry{ggplot2} Wizardry
{ggplot2} Wizardry
Переглядів 4,6 тис.3 роки тому
The recording from UseR Oslo's meetup 26/03/2021 www.meetup.com/Oslo-useR-Group/events/276513178/ In this talk, Cédric presented his favorite tips and tricks with regard to the {ggplot2} package, covering {ggplot2} functionality to improve the design of your graph as well as a collection of interesting extension packages. He covered a diverse collection of tips so hopefully everyone, no matter ...
{ggstatsplot}: An R Package for {ggplot2}-Based Plots With Statistical Details{ggstatsplot}: An R Package for {ggplot2}-Based Plots With Statistical Details
{ggstatsplot}: An R Package for {ggplot2}-Based Plots With Statistical Details
Переглядів 10 тис.3 роки тому
The recording from UseR Oslo's meetup on February 18th, 2021 - www.meetup.com/Oslo-useR-Group/events/275861577/ {ggstatsplot} is an extension of the {ggplot2} package for creating graphics with details from statistical tests included in the information-rich plots themselves. In a typical exploratory data analysis workflow, data visualization and statistical modeling are two different phases: vi...

КОМЕНТАРІ

  • @mkklindhardt
    @mkklindhardt 3 дні тому

    Very educational and still very high-level. Is there guiding videos/screencasts on meta-analysis with different types of response variables, modulators and outcomes? Perhaps covering the considerations of multivariate and multilevel meta-analysis as well as composite outcome meta-analyses, such as the concept of "borrowing of strength" (BoS) in meta-analysis where one has to deal with multiple outcomes (response variables) or studies with small sample sizes. Thank you

  • @davidanayonnaji3359
    @davidanayonnaji3359 13 днів тому

    Thank you for this insightful exposition to mirt. I am currently on an academic thesis at the University of Nigeria, Nsukka to study the DIF if a state examination, using many methods of which mirt is one. I find this video very instructive and followed through until I couldn't view the script you engaged to perform DIF. Could you share the script with me, or guide me to write my own script uniquely? Here is my unique case: I coded a 41-column dichotomously scored dataset with the name ESUPE _G.csv with the first column named gender as the grouping variable. In my data columns 2 to 40 are the item responses, while column 1 holds the gender grouping. I intend to run a DIF analysis using mirt package in R. How do I write a good R-script for the DIF analysis to flag DIF items? Thanks

  • @Featherlicht
    @Featherlicht 3 місяці тому

    One of the most useful videos that exists on the internet. Thank you for this.

  • @45tanviirahmed82
    @45tanviirahmed82 4 місяці тому

  • @user-xu7pg8fw9q
    @user-xu7pg8fw9q 4 місяці тому

    Thank you very much for the video, it is very helpful! I am glad I have a chance to ask you a question here about it as well! I have the following problem. It happened that I have only one level-2 variable and it has several missing values(it was an online survey from two organizations and there was a trouble with the data collection, so I am am sure the missing data is MCAR). I try to use 2lonly.pmm to fill in the gaps. I specify this variable as a -2 in pred: suppressWarnings(suppressMessages( ini <- mice(df, maxit = 0) )) pred <- ini$pred pred [,"school"] <- -2 When running the mice function it gives me the following error: Error in check.cluster(data, predictorMatrix) : Convert cluster variable school to integer by as.integer() I switched the organization variable to an integer and the error message then became as the following: Error in .imputation.level2(y = y, ry = ry, x = x, type = type, wy = wy, : No class variable I added additional variable "city", where both organizations are at, and specified it as a cluster variable in the pred. It didn't help, same error. Then I tried to specify the full pred matrix in excel and read it from a csv file as a matrix, the error changed to: Error in predictorMatrix[h, ] : subgroup out of bounds I consulted your book again and saw the following passage: "It is conceptually straightforward to extend imputations to higher levels (Yucel 2008). If there are two levels, combine all level-2 predictors with an aggregate (e.g., the cluster means) of the level-1 predictors and the level-1 outcomes. Once we have this, we may choose suitable methods from Chapter 3 to impute the missing level-2 variables in the usual way. No new issues arise." Is there a way to fix my problem? Or, given the above, is there maybe a way to not treat this variable as a cluster variable and just impute it with logreg in mice since it's the only one variable? Do you mean by aggregate to compute mean values or cluster variables? Thank you in advance for reading the post and have a great day ahead! Best Regards, Yulia

  • @mikiallen7733
    @mikiallen7733 6 місяців тому

    definitely a step in the right direction that can't come by chance from a neuroscientist like you , thank you very much

  • @duncanunwin3261
    @duncanunwin3261 6 місяців тому

    The best introduction to SEM I''ve seen. Thanks.

  • @fabiocarbone5342
    @fabiocarbone5342 6 місяців тому

    Thank you very much, very insightful lesson. Could you please tell me where I can also find a lesson or some material on meta-analysis of survival data (Hazard Ratios). It would really help. Thanks

  • @fowobajek
    @fowobajek 7 місяців тому

    Just what I have been looking for !. Thanks a lot.

  • @fabianrosenthal4892
    @fabianrosenthal4892 7 місяців тому

    What a great and insightful talk! Thanks a lot🎉

  • @Emir_-ws2yw
    @Emir_-ws2yw 8 місяців тому

    Hi. My first question for MetaSem: do the correlation coefficients have to be given in the primary studies? For example, can't MetaSem be done using d or g effect sizes? My other question is what exactly is the difference between MetaSem and meta-analysis? Thank you very much in advance for your answers...

  • @xuyang2776
    @xuyang2776 10 місяців тому

    Hello, Author. Could you tell me how to get the residual vairances of a MSE by lavaan()? Thanks

  • @belindadavey3529
    @belindadavey3529 10 місяців тому

    What about plotting posterior distributions and seeing how well they mirror the observed distribution (ie the PPE check)? Similarly, plotting predicted vs observed scatterplots, like in Shinystan for HLM models? Ty :)

  • @vidslearnig
    @vidslearnig Рік тому

    can you just show me how to open metasem in R studio

  • @giacomoiezzi9294
    @giacomoiezzi9294 Рік тому

    keep up the good work mate!

  • @_inspirationandmotivation

    Thank you!:)

  • @meseretmuche6984
    @meseretmuche6984 Рік тому

    Really fascinating video, please share me if you have a video lecture on the continuous data meta analysis for ecological study

  • @MrCigarro50
    @MrCigarro50 Рік тому

    Thank you for such a great video.

  • @LuisAguilar-pi9nq
    @LuisAguilar-pi9nq Рік тому

    Thank you for the package, is very helpful.

  • @tomatolz455
    @tomatolz455 Рік тому

    It's a great great course. Thank you!

  • @amiraalaamri7032
    @amiraalaamri7032 Рік тому

    Thanks a lot for this great presentation. It is very informative.

  • @scarlettzhang7800
    @scarlettzhang7800 Рік тому

    Thank you very much to all presenters! This is a very newbie-friendly video to learn SEM using R! Everything is well organized and well explained! Looking for more good videos:)

  • @Procrastinus
    @Procrastinus Рік тому

    Teaching this to my class thanks to this. Much thanks!

  • @miao9732
    @miao9732 Рік тому

    thank you so much for this video and other lectures of this channel!!!

  • @nabilataleb3132
    @nabilataleb3132 Рік тому

    Please more. Explanation of item caractéristique curve in rash model

  • @Kris-xy3kh
    @Kris-xy3kh Рік тому

    Thank you for this video! Do you have any advice on how to organzie data when doing the extracting from studies? Do you use excel or something? I want to do a meta analysis on low birth weight and mental health, but there are sooo many different tests for mental health that I cant think out a smart way to organize it in excel.

    • @UseROslo
      @UseROslo Рік тому

      Hi, this channel only focuses on publishing content from the UseR Oslo group. For specific questions about the content, we encourage you to contact the speaker or relevant forums.

  • @klaows
    @klaows Рік тому

    Why we need to use absolute latitude? Can I change to other variables?

    • @UseROslo
      @UseROslo Рік тому

      Hi, this channel only focuses on publishing content from the UseR Oslo group. For specific questions about the content, we encourage you to contact the speaker or relevant forums.

  • @klaows
    @klaows Рік тому

    Thank you, What's Ablat? How could I get Ablat?

  • @CocoFidarella
    @CocoFidarella Рік тому

    Thank you for the package! It's pretty good! I'm wondering, if you have a citeable source for thresholds of epsilon²? I need to interpret the effect size but can not find any source for thresholds or a rule of thumb for the interpretation... Any help is appreciated.

    • @UseROslo
      @UseROslo Рік тому

      Hi, this channel only focuses on publishing content from the UseR Oslo group. For specific questions about the content, we encourage you to contact the speaker or relevant forums.

  • @yeeunchoi3770
    @yeeunchoi3770 Рік тому

    Thank you so much for the video. It helped me a lot to understand and interpret the results of blavaan!

  • @iirolenkkari9564
    @iirolenkkari9564 Рік тому

    Amazing package. Thank you! I'm wondering how to model the covariance structure in a bayesian longitudinal setting, similar to covariance patterns such as compound symmetry, autoregressive, Topelitz etc. in the frequentist world. In the frequentist world, taking serial correlation into consideration narrows the confidence intervals of the parameters. How to model the covariance structure in a bayesian longitudinal setting? I'm wondering if a bayesian intercept always introduces compound symmetry, similar to a random intercept in a frequentist linear mixed effects model? I suspect taking serial correlation would narrow the posterior distributions of the model parameters, strengthening the bayesian inference. However, I'm not at all sure if my thoughts are anywhere near correct. The brms package is a very valuable resource. However, the parts about covariance structures seem to be still in progress. If anyone has good theoretical (and why not practical) bayesian references regarding these covariance modeling issues (serial correlation etc.), I would appreciate them very much.

    • @UseROslo
      @UseROslo Рік тому

      Hi, this channel only focuses on publishing content from the UseR Oslo group. For specific questions about the content, we encourage you to contact the speaker or relevant forums.

  • @clementinetine5355
    @clementinetine5355 Рік тому

    Thank you so so so much for this AMAZING tutorial!!! This is actually my second time trying to use R, and this video was a huge help in understanding the basic logic behind conducting SEM using R. I could even code the mediation model by myself in 3 minutes, which made me very happy :D You are a great educator

  • @prabpharm07
    @prabpharm07 2 роки тому

    Thank you for your service to the descipline of evidence synthesis.

  • @leonelemiliolereboursnadal6966
    @leonelemiliolereboursnadal6966 2 роки тому

    Many thanks for developing a package lile this

  • @danielforster4804
    @danielforster4804 2 роки тому

    The caretList function doesn't seem to exist.

    • @UseROslo
      @UseROslo 2 роки тому

      Most likely replaced in a newer version. Version 6.0 came out in April, and could not find the function there. I will reach out to the presenter to see if caretList has been replaced by a similarly working function.

    • @danielforster4804
      @danielforster4804 2 роки тому

      @@UseROslo I'm also wondering what are the nonparametric options for caret? The package is generally pretty easy to use but my data is completely impossible to normalise so it's either use nonparametrics or bootstrap it all to kingdom come.

    • @anitawen7165
      @anitawen7165 Рік тому

      I added the following libraries and it worked. library(caret) library(caretEnsemble)

  • @SagangaKapaya
    @SagangaKapaya 2 роки тому

    Thank you, great work and it helps a lot, inspires me to apply the method in my research works.

  • @martinwatts6362
    @martinwatts6362 2 роки тому

    Are you able to extract model fit values, such as AIC and log-likelihood, after applying the coef_test function? Thanks.

    • @UseROslo
      @UseROslo 2 роки тому

      Hi Martin. I recommend you reach out to the speaker with your question, or check his webpage: www.jepusto.com/

    • @martinwatts6362
      @martinwatts6362 2 роки тому

      @@UseROslo Thanks. Will do !

  • @ahmetemin1721
    @ahmetemin1721 2 роки тому

    Definitely one of the best videos about machine learning. It was quite informative and the narration was smooth. Thank you very much 😃

  • @hesamghiasvand1643
    @hesamghiasvand1643 2 роки тому

    So insightful and applicable. Many thanks. Also would be very grateful if you could provide such videos for Network Meta-Analysis. Thanks.

    • @UseROslo
      @UseROslo 2 роки тому

      There are no such talks planned at the moment, but will keep the topic in mind.

  • @joaoschneider4822
    @joaoschneider4822 2 роки тому

    30:27 Intro 34:38 Speakers Talk

    • @UseROslo
      @UseROslo 2 роки тому

      Thanks! An edited version will be put out soonish.

  • @yousif533
    @yousif533 2 роки тому

    How to change the font style of part of a string in forestplot? Specifically, I want to italicize the text (ABCX (rs1111111-X)), which is the name of one of the columns.

    • @UseROslo
      @UseROslo 2 роки тому

      Hi Yousif, I can't find anything in the metafor documentation that allows you to do that. I would recommend you check out relevant forums for metafor, or contact the package creator.

  • @SwedishRagers
    @SwedishRagers 2 роки тому

    Is it possible to forest plot odds ratios when only given the ratio and upper and lower limits? Since many studies plainly list their OR with 95%CI and p value without reporting the samples sizes. I cannot seem to find any videos regarding this and would greatly appreciate some advice.

    • @UseROslo
      @UseROslo 2 роки тому

      Hi Samuel. I recommend that you check the relevant forums for metafor. Wolfgang's website might be a good place to start: www.metafor-project.org/

    • @SwedishRagers
      @SwedishRagers 2 роки тому

      @@UseROslo Thank you very much for the timely response. And for all your quality content. We all greatly appreciated it!

  • @sultanmohammadahsan9521
    @sultanmohammadahsan9521 2 роки тому

    Wonderful package and very simple syntax...

  • @BedouinTravels
    @BedouinTravels 2 роки тому

    Thank you for this useful video! Can you please talk about experimental meta-analysis using R

    • @UseROslo
      @UseROslo 2 роки тому

      Unfortunately we do not have any such talks planned for Oslo UseR. You can see announcements for future events here: twitter.com/ourmeetup

  • @ogbucharles4539
    @ogbucharles4539 2 роки тому

    Hi, I am struggling to run a metanalysis. It is my first time. I have continuous data with mean, sample size, and standard deviation for each study (without means for control and exposure groups.) Any ideas on the appropriate argument suitable for this type of data?

    • @UseROslo
      @UseROslo 2 роки тому

      Hi, this is outside of our scope of knowledge, as the channel is run by the Oslo UseR organizers, and the individual speakers do not moderate the recordings we upload. Please refer to relevant forums or metafor's own webpage, www.metafor-project.org/

  • @jayh2247
    @jayh2247 2 роки тому

    Wonderful presentation! Thank you so much! :)

  • @juiceorca7528
    @juiceorca7528 2 роки тому

    Very useful, good video!

  • @TheStatsDon
    @TheStatsDon 2 роки тому

    Thank you

  • @sempredev7703
    @sempredev7703 2 роки тому

    Really great meetup, a little question here, why do we need RVE when the multilevel model is already designed for dependent effect sizes at a certain level?

    • @UseROslo
      @UseROslo 2 роки тому

      Hi, this channel is primarily for posting the recordings, and your questions goes beyond my knowledge unfortunately. I recommend you contact the speaker directly with your question.

    • @ST-fl9ek
      @ST-fl9ek Рік тому

      Hi, I got the same question too, have you got an answer so far? I think multilevel models assume cross-level residuals independence, unless clubSandwich adjusts for that?

  • @claire2247
    @claire2247 2 роки тому

    Very useful tutorial, thanks you!!