Nathaniel Haines

Nathaniel Haines

@Nate__Haines

Followers852
Following605

Clinical psych PhD candidate interested in externalizing psychopathology, computational/cognitive modeling, reinforcement learning, Bayes, and #rstats (he/him)

Columbus, OH
Joined on September 10, 2016
Statistics

We looked inside some of the tweets by @Nate__Haines and here's what we found interesting.

Inside 100 Tweets

Time between tweets:
5 hours
Average replies
49
Average retweets
306
Average likes
1145
Tweets with photos
32 / 100
Tweets with videos
5 / 100
Tweets with links
0 / 100

Computational models are not always easy to understand and for that reason are often disregarded. Sometimes, however, what appears to be a complex formal/algorithmic description encapsulates simple principles, in a coherent and unified account. https://t.co/3D3EWOj1vD

📢 Excited to be able to finally share this tutorial for Social Psychology: "Formalizing verbal theories: A tutorial by dialogue" by @MarkBlokpoel and myself. 🧵👇 1/n https://t.co/4tcygU4x4j

Really pleased that this paper has now been accepted for publication in Psychological Medicine! Double-celebrations tonight as this was a collaboration with my wife. First work from my lab applying computational modelling to clinical questions. https://t.co/ldhi5EAdQp

5/5 From this perspective, increasing the number of different items to ensure high reliability is just a heuristic. What really matters is that we can estimate data-generating parameters (or "true scores", see https://t.co/sXmUc0m0sO) with enough precision for a given inference.

4/5 The take-away being that we can build models of the temporal dynamics of whatever behavior we are interested in, and we can use simulations to determine if we can reliably estimate parameters from these person-specific models given the number of observations we have, etc.

3/5 And then we can try estimating the parameters from the simulated data to determine if we can accurately, and precisely, recover the "true" parameters. This is akin to assessing the reliability of our measure, but before we have even collected empirical data. https://t.co/hm3dADn3FF

3/5 And then we can try estimating the parameters from the simulated data to determine if we can accurately, and precisely, recover the "true" parameters. This is akin to assessing the reliability of our measure, but before we have even collected empirical data. https://t.co/hm3dADn3FF

2/5 The basic idea being that we can model learning dynamics using observed preferences among two choices, each with different average payoffs (1 choice is better on average, but this must be learned). We can simulate behavior with these models: https://t.co/AdcPe9Oozl https://t.co/wuJe900Eny

2/5 The basic idea being that we can model learning dynamics using observed preferences among two choices, each with different average payoffs (1 choice is better on average, but this must be learned). We can simulate behavior with these models: https://t.co/AdcPe9Oozl https://t.co/wuJe900Eny

1/5 One example of modeling changes in "a single item" over time is given by reinforcement learning. This series of blogs details a simple example: https://t.co/TXNx1eQN8N https://t.co/pXefhOXGnY https://t.co/sBkYq1GUaA

1/5 One example of modeling changes in "a single item" over time is given by reinforcement learning. This series of blogs details a simple example: https://t.co/TXNx1eQN8N https://t.co/pXefhOXGnY https://t.co/sBkYq1GUaA

Quoted @betanalpha

Y’all. I’ve been doing math for the last seven months with some excellent people. Look forward to another theory paper that no one will read but many will claim to have read. https://t.co/sRVuT6MONh

Theory seems to be under-appreciated in many areas of science. Everyone agrees that theory is important, but our publishing incentives push us toward empiricism... https://t.co/C8f8sy5DyZ

I just wrote a brief tutorial on how to avoid common issues and speed things up when doing row-wise operations in a data frame using dplyr. Check it out! #rstats #r4ds #tidyverse 🔗 https://t.co/cB7ZRn7SxM https://t.co/XAraPkuHnX

I just wrote a brief tutorial on how to avoid common issues and speed things up when doing row-wise operations in a data frame using dplyr. Check it out! #rstats #r4ds #tidyverse 🔗 https://t.co/cB7ZRn7SxM https://t.co/XAraPkuHnX

Universities filing amicus briefs/lawsuits against new ICE regs by my count: - Cornell - Dartmouth - Georgetown - Harvard - Hopkins (forthcoming) - Michigan State - MIT - Northeastern - Northwestern - Penn - Princeton - Purdue - Stanford - U of California - UMich - USC - Yale

Proof for the doubters of this glorious feat https://t.co/J2hZLSqBX5

Proof for the doubters of this glorious feat https://t.co/J2hZLSqBX5

Quoted @PeteEtchells

This flood of papers irritates me, in part because I’ve been on the receiving end of emails from this guy in the past where he basically shouted me down because I don’t have as many publications as him. https://t.co/nj1L4l1IAN

This flood of papers irritates me, in part because I’ve been on the receiving end of emails from this guy in the past where he basically shouted me down because I don’t have as many publications as him. https://t.co/nj1L4l1IAN

"just had the 757th paper of my academic career published this morning" That's nothing—I just published my 2,234th tweet. https://t.co/CPsw42Wn2O

The virtual venue for MathPsych/ICCM will activate soon! Presenters will be contacted by email with login information in the next few days so they can upload their recorded presentations this weekend. Is your presentation already good to go? Drop us a note to get early access.

just so we're clear, these are the types of people who have been driving this 'cancel culture' whining. Academia, pleeeeaasseee stop falling for it. https://t.co/m6svgFybBd

just so we're clear, these are the types of people who have been driving this 'cancel culture' whining. Academia, pleeeeaasseee stop falling for it. https://t.co/m6svgFybBd

New preprint out with @sadieonscience, Zöe Fenn, @GregoryRSL and @rui__mata. Temporal Discounting Across Adulthood: A Systematic Review and Meta-analysis https://t.co/Vw9BEPMxoB via @OSFramework. @CVLneuro @BBSutdallas @DukeBrain 1/8

Does anyone know of open datasets of risky-decision making across the lifespan? By risky DM, I mean tasks of the nature: "$3 with Pr(1) or $4 with Pr(.8), 0 otherwise".

Hi stats tweeps, any recommendations for good stats reference books to spend some grant money on? I'm looking for something on multilevel models, SEM, missing data, and GLM (categorical and count) in particular!

The ideas in the @Harpers letter were destroyed on Twitter yesterday. Here's a thread/meta-thread.

Next Page