Christopher Jarvis
@christopherjarvis.bsky.social
Operations Humanitarian Epidemiologist Researcher
https://www.linkedin.com/in/c-jarvis/
https://github.com/jarvisc1
Trustee @Mapaction
@AppliedEpi
#Rstats #Statssky #Episky
https://www.linkedin.com/in/c-jarvis/
https://github.com/jarvisc1
Trustee @Mapaction
@AppliedEpi
#Rstats #Statssky #Episky
Pinned
Rule of three (statistics) - Wikipedia
en.m.wikipedia.org
Looks fun here. Okay my fav thing in stats. How do you get an upper bound for a confidence interval when you have no events? 3/n
The rule of three. en.m.wikipedia.org/wiki/Rule_of...
The rule of three. en.m.wikipedia.org/wiki/Rule_of...
Reposted by Christopher Jarvis
Happy to share the publication of our work "Estimating social contact rates for the COVID-19 pandemic using Google mobility and pre-pandemic contact surveys", huge thanks to my co-authors @christopherjarvis.bsky.social @ngdavies.bsky.social @pietro-coletti.bsky.social Jantien Backer and John Edmunds
Redirecting
doi.org
April 28, 2025 at 7:46 AM
Happy to share the publication of our work "Estimating social contact rates for the COVID-19 pandemic using Google mobility and pre-pandemic contact surveys", huge thanks to my co-authors @christopherjarvis.bsky.social @ngdavies.bsky.social @pietro-coletti.bsky.social Jantien Backer and John Edmunds
Reposted by Christopher Jarvis
Stata users: if you want to make ‘kmunicate’-style Kaplan–Meier graphs, I’ve just released a Stata package on GitHub.
To get it:
. net from raw.githubusercontent.com/tpmorris/kmu...
User feedback welcome!
1/
To get it:
. net from raw.githubusercontent.com/tpmorris/kmu...
User feedback welcome!
1/
January 17, 2025 at 1:58 PM
Stata users: if you want to make ‘kmunicate’-style Kaplan–Meier graphs, I’ve just released a Stata package on GitHub.
To get it:
. net from raw.githubusercontent.com/tpmorris/kmu...
User feedback welcome!
1/
To get it:
. net from raw.githubusercontent.com/tpmorris/kmu...
User feedback welcome!
1/
This is a nice option that moves beyond lots of copy paste in your script and gets you closer to thinking about repeated patterns that you might eventually put in a package. #rstats
🎄Day 21: Ggplot efficiency part 3🎄
Need many plots with the same style/formatting but slightly different inputs?
💡Create a custom ggplot function!
Define your plot upfront in the function, with all the long formatting code. Then run the function for each new plot with different inputs.
Need many plots with the same style/formatting but slightly different inputs?
💡Create a custom ggplot function!
Define your plot upfront in the function, with all the long formatting code. Then run the function for each new plot with different inputs.
December 22, 2024 at 8:28 AM
This is a nice option that moves beyond lots of copy paste in your script and gets you closer to thinking about repeated patterns that you might eventually put in a package. #rstats
My fav shortcut is alt+shift+k which shows you all the shortcuts.
🎄Day 17: RStudio magic🎄
Want to be in multiple places at once over the holidays?
We can't help you with that.
But - we can show you how to type in multiple places at once in RStudio! 😮
Useful when you need to make the same edit across your code but Find + Replace isn't suitable.
Want to be in multiple places at once over the holidays?
We can't help you with that.
But - we can show you how to type in multiple places at once in RStudio! 😮
Useful when you need to make the same edit across your code but Find + Replace isn't suitable.
December 17, 2024 at 12:33 PM
My fav shortcut is alt+shift+k which shows you all the shortcuts.
I found #docker mysterious until I thought of it like buying a laptop setting it up exactly how I need and then handing it over to another person. Then dockerhub is like the post office that takes my laptop and delivers it to others. Tugboat makes it pretty easy to do in #rstats
December 17, 2024 at 12:19 PM
I feel like I started my #stats journey wanting to do lots of fancy multi-level, spatial, and Bayesian models. Then I'm now happy if I can produce a useful count, that I understand and can clearly explain.
In many ways knowing the complex stuff just gave me permission to do simple things. #rstats
In many ways knowing the complex stuff just gave me permission to do simple things. #rstats
December 17, 2024 at 10:48 AM
Glad to see @mapaction.bsky.social is now on Bluesky #gis #humanitarian #emergencyresponse #informationmanagement
December 10, 2024 at 7:56 PM
Glad to see @mapaction.bsky.social is now on Bluesky #gis #humanitarian #emergencyresponse #informationmanagement
I was chatting with a colleague about their #PhD viva on #spatialstats and I showed them this diagram that I made as as part of my thesis to help me understand the building blocks of #spatial models, especially for #inla. #statssky. pg 254 tinyurl.com/3ymp26p3
December 10, 2024 at 2:55 PM
I was chatting with a colleague about their #PhD viva on #spatialstats and I showed them this diagram that I made as as part of my thesis to help me understand the building blocks of #spatial models, especially for #inla. #statssky. pg 254 tinyurl.com/3ymp26p3
Today I learned that the R in #RShiny is doing a lot of the heavy lifting in that hashtag. #shinyisnotjustR #Rstats
December 9, 2024 at 6:34 PM
Today I learned that the R in #RShiny is doing a lot of the heavy lifting in that hashtag. #shinyisnotjustR #Rstats
Reposted by Christopher Jarvis
Knuth, one of the greats in CS history, famously said, "Premature optimization is the root of all evil...in programming." Only a small % of R apps need to be efficient. But they all need to be correct, and often loops etc. are clearer and thus more likely to be correct. Anti-loop mania is bad for R.
December 9, 2024 at 4:41 PM
Knuth, one of the greats in CS history, famously said, "Premature optimization is the root of all evil...in programming." Only a small % of R apps need to be efficient. But they all need to be correct, and often loops etc. are clearer and thus more likely to be correct. Anti-loop mania is bad for R.
bluesky is inspiring! I think I've made my first #R package #rstats #eda github.com/jarvisc1/qex.... Inspired by [email protected] and a chat with @pete-mck.bsky.social bsky.app/profile/did:.... I thought why not try to bring what I like about #Stata and #data.table to the #tidyverse.
December 8, 2024 at 3:37 PM
bluesky is inspiring! I think I've made my first #R package #rstats #eda github.com/jarvisc1/qex.... Inspired by [email protected] and a chat with @pete-mck.bsky.social bsky.app/profile/did:.... I thought why not try to bring what I like about #Stata and #data.table to the #tidyverse.
Something I find useful from base #R is the dput command.
When I have an object but I want to recreate it and copy into my script I chuck it into dput and it gives me the code that could be used to create it.
#Rstats #Datascience
When I have an object but I want to recreate it and copy into my script I chuck it into dput and it gives me the code that could be used to create it.
#Rstats #Datascience
December 7, 2024 at 3:24 PM
Something I find useful from base #R is the dput command.
When I have an object but I want to recreate it and copy into my script I chuck it into dput and it gives me the code that could be used to create it.
#Rstats #Datascience
When I have an object but I want to recreate it and copy into my script I chuck it into dput and it gives me the code that could be used to create it.
#Rstats #Datascience
Reminds me that there is a difference between being right and not having to be too wrong. It's great when the margin of error for success is wide. I also avoid absolutes! 'No one knows anything' or 'everyone is faking it' and I realise I'm very lucky to have met people with deep understanding.
Why ‘everyone gets it wrong sometimes’ is a bad argument when assessing track records for incoming leaders in science and health.
My new post: kucharski.substack.com/p/everyone-g...
My new post: kucharski.substack.com/p/everyone-g...
December 6, 2024 at 8:23 AM
Reminds me that there is a difference between being right and not having to be too wrong. It's great when the margin of error for success is wide. I also avoid absolutes! 'No one knows anything' or 'everyone is faking it' and I realise I'm very lucky to have met people with deep understanding.
I love this! Makes me think about stuff by Paul Graham and how clutter might relate to energy and perhaps (big perhaps!) kids have coarser perception so less drained. Maybe some of these high-energy rulers are Peter Pan like and never grew up? 😂 paulgraham.com/stuff.html
December 6, 2024 at 8:06 AM
I love this! Makes me think about stuff by Paul Graham and how clutter might relate to energy and perhaps (big perhaps!) kids have coarser perception so less drained. Maybe some of these high-energy rulers are Peter Pan like and never grew up? 😂 paulgraham.com/stuff.html
Reposted by Christopher Jarvis
I am increasingly convinced that besides the usual suspects like beauty, intellect and charisma, it's the people with energy who run the world
December 5, 2024 at 5:32 PM
I am increasingly convinced that besides the usual suspects like beauty, intellect and charisma, it's the people with energy who run the world
Reposted by Christopher Jarvis
Hello Bluesky! We’re Applied Epi! What do we want? To improve Applied Epidemiology worldwide! When do we want it? Well probably in an achievable timeframe that doesn’t completely destroy a much needed and already existing infrastructure that is overburdened!
December 5, 2024 at 11:39 PM
Hello Bluesky! We’re Applied Epi! What do we want? To improve Applied Epidemiology worldwide! When do we want it? Well probably in an achievable timeframe that doesn’t completely destroy a much needed and already existing infrastructure that is overburdened!
Reposted by Christopher Jarvis
ℹ️ Applying for postdoc roles in the UK? Check the helpful application advice thread by @rozeggo.bsky.social👇 #AcademicSky
🧪 Another great community contribution from Roz is IDDjobs!
🔗 Visit iddjobs.org for job opportunities in topics broadly aligned with infectious disease dynamics #IDsky #EpiSky
🧪 Another great community contribution from Roz is IDDjobs!
🔗 Visit iddjobs.org for job opportunities in topics broadly aligned with infectious disease dynamics #IDsky #EpiSky
A thread🧵 on application systems for postdoc jobs in the UK (applies to my university, but I've seen others and they're similar).
I have seen (too many times) good applicants who don't seem to understand how recruitment is done on our side - not their fault! it's a privilege! - so here's how
I have seen (too many times) good applicants who don't seem to understand how recruitment is done on our side - not their fault! it's a privilege! - so here's how
December 5, 2024 at 11:00 PM
ℹ️ Applying for postdoc roles in the UK? Check the helpful application advice thread by @rozeggo.bsky.social👇 #AcademicSky
🧪 Another great community contribution from Roz is IDDjobs!
🔗 Visit iddjobs.org for job opportunities in topics broadly aligned with infectious disease dynamics #IDsky #EpiSky
🧪 Another great community contribution from Roz is IDDjobs!
🔗 Visit iddjobs.org for job opportunities in topics broadly aligned with infectious disease dynamics #IDsky #EpiSky
I love this post. So for me and maybe for me only, the key is abstraction and context shifting. The more I can zoom in and out of different levels of abstraction without giving myself the burden of fully understanding at the moment the easier I can start to see patterns.
The hardest thing about public health training for me so far, is adjusting to being a generalist not a specialist.
My brain is a deep-diver and wants to have a solid grasp a topic.
Now I need to know just enough about lots of topics to get by. But all those topics change often. How do I keep up?
My brain is a deep-diver and wants to have a solid grasp a topic.
Now I need to know just enough about lots of topics to get by. But all those topics change often. How do I keep up?
December 5, 2024 at 10:51 PM
I love this post. So for me and maybe for me only, the key is abstraction and context shifting. The more I can zoom in and out of different levels of abstraction without giving myself the burden of fully understanding at the moment the easier I can start to see patterns.
This reminds me of a time I was sat in a meeting about efficiency savings where I was calculating the hourly cost of bringing everyone together being 10x any saving we could make. And we concluded that we would need several more meetings! 🤦♂️
Classic university inefficiency.
PhD students provide high quality teaching at low cost.
Financial pressures mean many universities are refusing to 'spend' money on PhD student teaching.
The work must therefore be done by expensive academics.
So you've reduced outgoings but increased costs!
PhD students provide high quality teaching at low cost.
Financial pressures mean many universities are refusing to 'spend' money on PhD student teaching.
The work must therefore be done by expensive academics.
So you've reduced outgoings but increased costs!
Teaching opportunities for PhDs dry up in university cash crisis
“Barely one in three postgraduate researchers (PGRs) are being given the opportunity to teach undergraduates, a major survey of UK and Australian PhD students has found.” [Click to read more]
“Barely one in three postgraduate researchers (PGRs) are being given the opportunity to teach undergraduates, a major survey of UK and Australian PhD students has found.” [Click to read more]
December 5, 2024 at 10:34 PM
This reminds me of a time I was sat in a meeting about efficiency savings where I was calculating the hourly cost of bringing everyone together being 10x any saving we could make. And we concluded that we would need several more meetings! 🤦♂️
I also quite like maximum likelihood theory, for me it’s just someone saying hey if you had a big bunch of numbers and you could only pick one to represent them all what would you go for?
December 5, 2024 at 10:24 PM
I also quite like maximum likelihood theory, for me it’s just someone saying hey if you had a big bunch of numbers and you could only pick one to represent them all what would you go for?
My third favourite thing about stats. Once you learn how to do really complex modelling and methods you can give yourself permission to just calculate a percentage and make a decision. Because you’ve probably realised all the complexity didn’t affect anything meaningfully.
December 5, 2024 at 10:15 PM
My third favourite thing about stats. Once you learn how to do really complex modelling and methods you can give yourself permission to just calculate a percentage and make a decision. Because you’ve probably realised all the complexity didn’t affect anything meaningfully.
My next fav thing in stats simulation studies. You really can’t go wrong by following @timpmorris.bsky.social and reading this onlinelibrary.wiley.com/doi/10.1002/...
It pretty much allows you to play God and understand why and when methods fail. Even when you’ve invented your own method!
It pretty much allows you to play God and understand why and when methods fail. Even when you’ve invented your own method!
Using simulation studies to evaluate statistical methods
Simulation studies are computer experiments that involve creating data by pseudo-random sampling. A key strength of simulation studies is the ability to understand the behavior of statistical methods....
onlinelibrary.wiley.com
December 5, 2024 at 10:12 PM
My next fav thing in stats simulation studies. You really can’t go wrong by following @timpmorris.bsky.social and reading this onlinelibrary.wiley.com/doi/10.1002/...
It pretty much allows you to play God and understand why and when methods fail. Even when you’ve invented your own method!
It pretty much allows you to play God and understand why and when methods fail. Even when you’ve invented your own method!
Looks fun here. Okay my fav thing in stats. How do you get an upper bound for a confidence interval when you have no events? 3/n
The rule of three. en.m.wikipedia.org/wiki/Rule_of...
The rule of three. en.m.wikipedia.org/wiki/Rule_of...
Rule of three (statistics) - Wikipedia
en.m.wikipedia.org
December 5, 2024 at 10:01 PM
Looks fun here. Okay my fav thing in stats. How do you get an upper bound for a confidence interval when you have no events? 3/n
The rule of three. en.m.wikipedia.org/wiki/Rule_of...
The rule of three. en.m.wikipedia.org/wiki/Rule_of...
This is how it works in the civil service except it’s binary. Take the bullet points, write your response in the same order as the advert, with the same key words. Maybe even keep the headings. Don’t try to be clever or original it won’t help you.
A thread🧵 on application systems for postdoc jobs in the UK (applies to my university, but I've seen others and they're similar).
I have seen (too many times) good applicants who don't seem to understand how recruitment is done on our side - not their fault! it's a privilege! - so here's how
I have seen (too many times) good applicants who don't seem to understand how recruitment is done on our side - not their fault! it's a privilege! - so here's how
December 5, 2024 at 9:56 PM
This is how it works in the civil service except it’s binary. Take the bullet points, write your response in the same order as the advert, with the same key words. Maybe even keep the headings. Don’t try to be clever or original it won’t help you.