What Drives Belief in Data Analysis: Stochastic or Fatalistic Ideologies?

  • Thread starter Thread starter shadowpuppet
  • Start date Start date
  • Tags Tags
    Data
Click For Summary

Discussion Overview

The discussion explores the ideologies influencing belief in data analysis, specifically contrasting stochastic and fatalistic perspectives. Participants examine various statistical frameworks, including Ætiologic, Bayesian, Frequentist, Nomothetic, Pyrrhonian, and Solipsist approaches, and how these relate to their personal methodologies in justifying conclusions.

Discussion Character

  • Debate/contested
  • Conceptual clarification
  • Exploratory

Main Points Raised

  • Some participants propose that stochastic ideologies favor high recurrence and correlativity, while fatalistic ideologies deduce future or past events from present circumstances.
  • One participant describes Ætiologic as isolating mechanisms for direct causation, while Bayesian includes extraneous a priori considerations.
  • Frequentist approaches are characterized by a reliance on empirical evidence and probability, with one participant noting the limitations of memory as a sample space.
  • Nomothetic reasoning is described as intuitive, though some participants express skepticism about its sole reliance.
  • Pyrrhonian skepticism is acknowledged as doubting all premises, including skepticism itself, while Solipsism is viewed as a philosophical consideration rather than a practical belief.
  • There is a suggestion that Bayesian and Frequentist approaches are not mutually exclusive, prompting further inquiry into their relationship.
  • One participant expresses doubt about the validity of Bayesianism in practical applications, favoring a more physicist-like perspective.
  • Concerns are raised about the political implications of certain statistical methods, particularly regarding the assumptions made in Bayesian analysis.

Areas of Agreement / Disagreement

Participants express a range of views on the ideologies and methodologies discussed, with no clear consensus emerging. Some agree on the definitions and implications of certain approaches, while others challenge or refine these ideas, indicating ongoing debate.

Contextual Notes

Participants highlight limitations in their understanding of certain terms and concepts, such as the relationship between Bayesianism and political motivations, and the role of memory in empirical analysis. There are also references to specific cognitive processes that may influence data interpretation.

Mode of Inference

  • Ætiologic - isolates mechanisms for direct causation

    Votes: 0 0.0%

  • Bayesian - tends to include extraneous a priori considerations

    Votes: 0 0.0%

  • Frequentist - respects only empirical evidence and probability

    Votes: 1 50.0%

  • Nomothetic - uses an intuitive system of diagnostic references

    Votes: 1 50.0%

  • Pyrrhonian - will doubt anything, even skepticism itself

    Votes: 0 0.0%

  • Solipsist - ascribes perceptual qualia exclusively to the mind

    Votes: 0 0.0%
  • Total voters
    2
shadowpuppet
Messages
30
Reaction score
0
Specifically, how do you come to believe what you do? Common methods of justifying novel premises epistemologically tend to favor one of two ideologies:

  • Stochastic - likes to interpret according to high recurrence and correlativity
  • Fatalist - believes that future/past can be deduced from knowledge of present circumstances

These are the two ideological premises from which the six poll methods can be derived: Ætiologic, Nomothetic, and Solipsist arguments can be seen primarily as assertions of determinism, whereas Bayesian, Frequentist, and Pyrrhonian frames of reference always adhere to a statistical approach. There is also another undercurrent running in this poll: Frequentist and Ætiologic justifications tend to employ an exclusively empirical underpinning; Nomothetic and Bayesian inferences depend heavily on a priori convictions (Solipsism is also usually defended using the a priori because a posteriori attempts at verification are not widely credited; Pyrrhonism may seem like an analytic proposition at first but it actually only indoctrinates an inductive negation of premises, including the self-negation of any premise that might eventually support a rationalist Pyrrhonian criterion, and so is actually an [anti-] empirical enterprise - I have taken great pains to clarify this in the past; you can hear my detailed arguments http://youtube.com/watch?v=mdEreZjrNeM").
 
Last edited by a moderator:
Physics news on Phys.org
I don't really justify much, I have a healthy mix of comfort and obligation in my work (i.e. I do what I want and what it takes to get by):

Ætiologic - isolates mechanisms for direct causation
Bayesian - tends to include extraneous a priori considerations
Frequentist - respects only empirical evidence and probability
Nomothetic - uses an intuitive system of diagnostic references
Pyrrhonian - will doubt anything, even skepticism itself
Solipsist - ascribes perceptual qualia exclusively to the mind

Ætiologic - don't use, I think it's politicians?
Bayesian - politicians...?

Frequentist - me at work and school and sometimes during introspection (your memory is a terrible sample space though, methinx, but maybe there's a reason some events stick out more than others, we call it 'weighting' in statistics, *chuckle*).

Nomothetic - I practice this a lot, sometimes even at work, but I never rely solely on it, and investigate as a Frequentist before.

Pyrrhonian - I go here sometimes.

Solipsist - it's an interesting point that has it's merits and I often consider it in a philosophical atmosphere, but I'm not a 'brain in a vat'.
 
Bayesian / Frequentist
 
Pythagorean said:
I don't really justify much, I have a healthy mix of comfort and obligation in my work (i.e. I do what I want and what it takes to get by):

Ætiologic - don't use, I think it's politicians?
Bayesian - politicians...?

Frequentist - me at work and school and sometimes during introspection (your memory is a terrible sample space though, methinx, but maybe there's a reason some events stick out more than others, we call it 'weighting' in statistics, *chuckle*).

Nomothetic - I practice this a lot, sometimes even at work, but I never rely solely on it, and investigate as a Frequentist before.

Pyrrhonian - I go here sometimes.

Solipsist - it's an interesting point that has it's merits and I often consider it in a philosophical atmosphere, but I'm not a 'brain in a vat'.

I agree with your description of solipsism, but I am not sure what you mean by 'politicians' and I think that the activation of memory has a lot to do with long-term potentiation and parallel connectivity in the hippocampus and cortex. However, if you primarily consider yourself to be a Frequentist, please remember to vote for it in the poll.

Moridin said:
Bayesian / Frequentist

Bayesian is a statistical form of rationalism and Frequentist is a statistical form of empiricism. Do you tend to prefer to theorize in your mind (Bayesian) or would you rather experiment in the real world (Frequentist)? Don't forget to vote!
 
I voted nomothetic because I am always thinking like a physicist, I dislike frequentist and Ætiologic deduction (although it reminds me of engineering), I doubt the validity of bayesianism applied to reality, and I favor Wittgenstein's rejection of Pyrrhonism and Solipsism.
 
Bayesian is a statistical form of rationalism and Frequentist is a statistical form of empiricism. Do you tend to prefer to theorize in your mind (Bayesian) or would you rather experiment in the real world (Frequentist)? Don't forget to vote!

I don't think they are mutually exclusive.
 
shadowpuppet said:
I agree with your description of solipsism, but I am not sure what you mean by 'politicians' and I think that the activation of memory has a lot to do with long-term potentiation and parallel connectivity in the hippocampus and cortex. However, if you primarily consider yourself to be a Frequentist, please remember to vote for it in the poll.

Yeah, my answers were a bit terse and I see a grammar error even. I meant that these seem like methods meant for making big conclusions that are inevitably at the mercy of politics.

Here's your definitions:

Ætiologic - isolates mechanisms for direct causation
Bayesian - tends to include extraneous a priori considerations

I might have a misunderstanding about Baeyesian, but showing causation is often associated with political motivation in my mind; extraneous a priori considerations reminds me of a large set of data that you've made assumptions about in order to arrive at your conclusion about causation.