Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I don't use Bayes factors in my research (2019)

datacolada.org

Post reply on HN
Print view View original

41–50 of 78 posts

Re: I don't use Bayes factors in my research (2019)

#41

Post by kkoncevicius »

civilized wrote: post #39
kkoncevicius wrote: post #37

Earlier quoted context omitted.

> Bayesian methods are not easy to use, I agree. But that's because they're trying to answer much more meaningful (but harder) questions than frequentist ones I think your view about the difference between Frequentist and Bayesian methods is wrong. There is this rant I like from Larry Wasserman [1] on the subject: My opinions have shifted a bit. [...] Bayes-Frequentist debate still matters. And people — including man…

Having a specific opinion isn't inherently a bias, and it's rather uninspiring to put effort into a substantial comment and then get a reply which is essentially nothing but a baseless accusation of bias followed by a long block quote of unclear relevance (it's extremely far from true that all statistical analyses worth consideration are in high dimensions). Review the Hacker News guidelines: > Edit out swipes. https…

Hm, that line doesn't seem bad to me. But I can certainly edit it if you think it's harsh in some way. Which would you prefer?:

  1. "I think your view about the difference between Frequentist and Bayesian methods is biased"
  2. "I think your view about the difference between Frequentist and Bayesian methods is wrong".
  3. Any other suggestion?

Re: I don't use Bayes factors in my research (2019)

#42

Post by SiempreViernes »

kkoncevicius wrote: post #37
civilized wrote: post #34

Bayesian methods are not easy to use, I agree. But that's because they're trying to answer much more meaningful (but harder) questions than frequentist ones, which researchers should be trying to do. You can't ignore Bayesian epistemology just by not using Bayesian methods. The underlying considerations in the Bayesian framework will inevitably become relevant to how you interpret your data, whether or not you use a…

> Bayesian methods are not easy to use, I agree. But that's because they're trying to answer much more meaningful (but harder) questions than frequentist ones I think your view about the difference between Frequentist and Bayesian methods is wrong. There is this rant I like from Larry Wasserman [1] on the subject: My opinions have shifted a bit. [...] Bayes-Frequentist debate still matters. And people — including man…

Honestly I can't tell how the quote is supposed to augment your point.

Re: I don't use Bayes factors in my research (2019)

#43

Post by travisjungroth »

constantcrying wrote: post #17
concinds wrote: post #11

Earlier quoted context omitted.

I read a survey once, that found that a huge number of PhDs/researchers in the studied sample gave an incorrect definition for what a "95% confidence interval" (/p-value, etc) actually means, and that several popular introductory textbooks defined it incorrectly as well. Wish I bookmarked it. At bare minimum, journals need to require that researchers publish all their data alongside every paper, so statistical analys…

I think you may be talking about "Mindless statistics" by Gigerenzer. He has some surveys about p-values and how radically wrong they are usually interpreted. >At bare minimum, journals need to require that researchers publish all their data alongside every paper, so statistical analyses can be redone and flaws can be spotted. Absolutely.

P-values work great when they’re super low, experiments run at a human-scale frequency, and hypotheses are extremely precise in their predictions, e.g. some physics.

If you run an experiment a day and get p interpretation of p-values approximately don’t matter. Running social sciences experiments with p < 0.05 threshold is where things get weird.

Re: I don't use Bayes factors in my research (2019)

#44

Post by constantcrying »

ggm wrote: post #33
constantcrying wrote: post #30

Earlier quoted context omitted.

It isn't an adhom because it is irrelevant.

This isn't a very good line of reasoning because it's a non sequiteur. At this point I think you're trolling more than discussing rationally.

It is an article about a statistical concept.

Re: I don't use Bayes factors in my research (2019)

#45

Post by civilized »

kkoncevicius wrote: post #41
civilized wrote: post #39

Earlier quoted context omitted.

Having a specific opinion isn't inherently a bias, and it's rather uninspiring to put effort into a substantial comment and then get a reply which is essentially nothing but a baseless accusation of bias followed by a long block quote of unclear relevance (it's extremely far from true that all statistical analyses worth consideration are in high dimensions). Review the Hacker News guidelines: > Edit out swipes. https…

Hm, that line doesn't seem bad to me. But I can certainly edit it if you think it's harsh in some way. Which would you prefer?: 1. "I think your view about the difference between Frequentist and Bayesian methods is biased" 2. "I think your view about the difference between Frequentist and Bayesian methods is wrong". 3. Any other suggestion?

I don't find it a helpful comment either way, but "wrong" would at least shed the implication of measuring unfairness from an objective standpoint.

Re: I don't use Bayes factors in my research (2019)

#46

Post by constantcrying »

travisjungroth wrote: post #43
constantcrying wrote: post #17

Earlier quoted context omitted.

I think you may be talking about "Mindless statistics" by Gigerenzer. He has some surveys about p-values and how radically wrong they are usually interpreted. >At bare minimum, journals need to require that researchers publish all their data alongside every paper, so statistical analyses can be redone and flaws can be spotted. Absolutely.

P-values work great when they’re super low, experiments run at a human-scale frequency, and hypotheses are extremely precise in their predictions, e.g. some physics. If you run an experiment a day and get p interpretation of p-values approximately don’t matter. Running social sciences experiments with p < 0.05 threshold is where things get weird.

Did you read the article?

>even your interpretation of p-values approximately don’t matter

"Small number means good" is not a sufficient working understanding of p-values for doing science.

Re: I don't use Bayes factors in my research (2019)

#47

Post by kkoncevicius »

civilized wrote: post #45
kkoncevicius wrote: post #41

Earlier quoted context omitted.

Hm, that line doesn't seem bad to me. But I can certainly edit it if you think it's harsh in some way. Which would you prefer?: 1. "I think your view about the difference between Frequentist and Bayesian methods is biased" 2. "I think your view about the difference between Frequentist and Bayesian methods is wrong". 3. Any other suggestion?

I don't find it a helpful comment either way, but "wrong" would at least shed the implication of measuring unfairness from an objective standpoint.

A Bayesian proponent wanting an objective standpoint ... Sorry, could not resist :) But I edited the comment. Sorry if it ruined your mood. I shared it with good intentions, not to pick a fight.

Re: I don't use Bayes factors in my research (2019)

#48

Post by kgwgk »

travisjungroth wrote: post #40
LudwigNagasena wrote: post #36

Earlier quoted context omitted.

I think there is some confusion going on. Nobody claims that there is the null hypothesis. I think you are fighting windmills. Let’s say you study P. You know that P belongs to the family 𝒫. For example, 𝒫 = {N(μ,σ²): μ∈ℝ, σ²>0}. To be aware of 𝒫 is a prerequisite to do any sort of testing. For example, to test H0: μ=0 vs H1: μ≠0. After all, a typical hypothesis test is just a likehood ratio test. What you don’t h…

> Nobody claims that there is the null hypothesis. Google “the null hypothesis”. If you mean null model, then I’m not fighting against anyone. We all agree which null model to use is a choice to be made. Otherwise, I’m not even sure what you’re trying to convince me of at this point. I’ll restate the essence of my first comment more concisely. Bayes factors are a method of model comparison. You take the ratio of marg…

> Choosing a null model for this purpose requires more assumptions than doing null hypothesis testing with frequentist statistics.

How so?

I could choose the same null model that predicts that the observation is distributed as, say, a standard Gaussian. What additional assumptions are required?

Re: I don't use Bayes factors in my research (2019)

#49

Post by civilized »

kkoncevicius wrote: post #47
civilized wrote: post #45

Earlier quoted context omitted.

I don't find it a helpful comment either way, but "wrong" would at least shed the implication of measuring unfairness from an objective standpoint.

A Bayesian proponent wanting an objective standpoint ... Sorry, could not resist :) But I edited the comment. Sorry if it ruined your mood. I shared it with good intentions, not to pick a fight.

Thank you, cheers :)

Re: I don't use Bayes factors in my research (2019)

#50

Post by travisjungroth »

kgwgk wrote: post #48
travisjungroth wrote: post #40

Earlier quoted context omitted.

> Nobody claims that there is the null hypothesis. Google “the null hypothesis”. If you mean null model, then I’m not fighting against anyone. We all agree which null model to use is a choice to be made. Otherwise, I’m not even sure what you’re trying to convince me of at this point. I’ll restate the essence of my first comment more concisely. Bayes factors are a method of model comparison. You take the ratio of marg…

> Choosing a null model for this purpose requires more assumptions than doing null hypothesis testing with frequentist statistics. How so? I could choose the same null model that predicts that the observation is distributed as, say, a standard Gaussian. What additional assumptions are required?

Mean and standard deviation up front, not from the sample. At least, that would go against my understanding of Bayes factors and how I’ve calculated them. You can do other stats, t-test for example, without declaring that up front.
Post reply on HN