Discover and read the best of Twitter Threads about #outliers
Most recents (4)
hello #psychology #stats twitter!
a while ago i promised some graphs on why #removing #outliers using a simple cut-off (eg, >2SDs) is a #bad idea
so that I can sleep at night again, here they are
tldr: DON'T (blindly) USE FIXED OUTLIER CUT-OFFS LIKE >2SD. EVER.
1/15
a while ago i promised some graphs on why #removing #outliers using a simple cut-off (eg, >2SDs) is a #bad idea
so that I can sleep at night again, here they are
tldr: DON'T (blindly) USE FIXED OUTLIER CUT-OFFS LIKE >2SD. EVER.
1/15
some twitter background:
2/15
2/15
for some reason, it's very common in #psychology to remove 'outliers' from data
most common way: exclude data more than two standard deviations from mean
we spend time & money collecting data, then throw 5% away
🤷
I don't know why, or where it's taught, but there it is
3/15
most common way: exclude data more than two standard deviations from mean
we spend time & money collecting data, then throw 5% away
🤷
I don't know why, or where it's taught, but there it is
3/15
Daily Bookmarks to GAVNet 08/03/2021 greeneracresvaluenetwork.wordpress.com/2021/08/03/dai…
Retina 'hardwired' to predict path of moving objects
medicalxpress.com/news/2021-08-r…
#PrimateRetina #NeuralCircuits #PathPrediction #MovingObjects
medicalxpress.com/news/2021-08-r…
#PrimateRetina #NeuralCircuits #PathPrediction #MovingObjects
Allen's Pause: The British Agricultural Revolution at Bay, 1740-1800
daviskedrosky.substack.com/p/allens-pause
#BritishHistory #AgrarianChange #IndustrialRevolution #LandUse #specialization #MarketFlexibility
daviskedrosky.substack.com/p/allens-pause
#BritishHistory #AgrarianChange #IndustrialRevolution #LandUse #specialization #MarketFlexibility
"Most of the things that differ from the past will shortly die. The more they differ, the sooner they will die."
@SimonDeDeo of SFI & @CarnegieMellon on the linear correlation of #novelty & #transience - aka, "Why you mostly listen to the same songs over and over" - at #CSSS19:
@SimonDeDeo of SFI & @CarnegieMellon on the linear correlation of #novelty & #transience - aka, "Why you mostly listen to the same songs over and over" - at #CSSS19:
"It's true that new things die. It's also true that is you want to be very very big, be new."
@SimonDeDeo of SFI & @CarnegieMellon on successful #novelty & hopeful #monsters in #evolution, #innovation, & #investment - aka, "Either be really boring, or be Beethoven" - at #CSSS19:
@SimonDeDeo of SFI & @CarnegieMellon on successful #novelty & hopeful #monsters in #evolution, #innovation, & #investment - aka, "Either be really boring, or be Beethoven" - at #CSSS19:
"#Revolutions are not just about ideas, they're about #information management. You can't just run a #revolution in a seminar room."
"The stranger your speech is, the more likely people are to just nod & move on."
@SimonDeDeo on topic modelling #revolutionary speech at #CSSS19:
"The stranger your speech is, the more likely people are to just nod & move on."
@SimonDeDeo on topic modelling #revolutionary speech at #CSSS19:
As a rookie podcaster looking to connect deeper with audience communities, it’s encouraging to see new podcast launches everyday and talks of them being the next big thing. Sharing some early, honest lessons
First of all, it looks like the quality of conversation and the core story still wins audiences. And that’s not because it’s produced in a podcasting format.
It seems to be working with niche, specialized communities of listeners, and not potentially a mass product at scale. I’m talking 10,000 listens per episode being a gold standard. Quality, full listens.
