*

Offline Pete Svarrior

  • e
  • Planar Moderator
  • *****
  • Posts: 8665
  • (◕‿◕✿)
    • View Profile
    • The Flat Earth Society
Re: Rowbotham experiment #9
« Reply #60 on: January 06, 2019, 01:37:42 PM »
Nowhere in any of this thread did I use the word 'causation' or its cognates.
I refer you to my previous point on meaningless semantics, and how effective they are as a distraction strategy.
Read the FAQ before asking your question - chances are we've already addressed it.
Follow the Flat Earth Society on Facebook and Twitter!


*mic stays stationary and earth accelerates upwards towards it*

Offline edby

  • *
  • Posts: 1066
    • View Profile
Re: Rowbotham experiment #9
« Reply #61 on: January 06, 2019, 01:43:17 PM »
Nowhere in any of this thread did I use the word 'causation' or its cognates.
I refer you to my previous point on meaningless semantics, and how effective they are as a distraction strategy.
Perhaps we could refer this dispute on a technical point to a referee here. If no one here I have plenty of academic colleagues who could act as such. The exam question is whether a correlation of 97.5% over a sample size of 54 requires an explanation. [EDIT] I am happy to increase the sample size, plus include the corrections in the 1879 edition (see above), and publish the evidence. Up to you.

By all means consign to 'angry ranting about correlation', but then we have the evidence that you have done so.
« Last Edit: January 06, 2019, 01:53:04 PM by edby »

*

Offline Pete Svarrior

  • e
  • Planar Moderator
  • *****
  • Posts: 8665
  • (◕‿◕✿)
    • View Profile
    • The Flat Earth Society
Re: Rowbotham experiment #9
« Reply #62 on: January 06, 2019, 02:08:51 PM »
Nah, I apologise about the AR posturing and I take it back. I strongly disagree with your point, and I do think it relies on fallacies, but I don't doubt your genuine intentions. From a moderation standpoint, there's nothing wrong with this thread.
Read the FAQ before asking your question - chances are we've already addressed it.
Follow the Flat Earth Society on Facebook and Twitter!


*mic stays stationary and earth accelerates upwards towards it*

Offline edby

  • *
  • Posts: 1066
    • View Profile
Re: Rowbotham experiment #9
« Reply #63 on: January 06, 2019, 06:16:58 PM »
I strongly disagree with your point, and I do think it relies on fallacies ...

I will try to explain better. We have a function f(h) -> v, where h is lighthouse height, v is visible distance, and we are trying to test ‘how well’ the function predicts the visible range given by observers.

Suppose that in scenario 1, our function returns a random number uniformly distributed between 0 and 45 miles, where 45 miles is the maximum range that any light can be seen due to atmospheric conditions. I think you agree that would not be a very useful function. It predicts nothing, because it is random, and that was my point earlier about randomness.

In scenario 2, by contrast, suppose that the maximum difference either side is 5 miles, that 80% of the observations are within 2 miles of the amount predicted by our function, and that 70% are within 1 mile.

Then we don’t need adjectives like ‘good’ or ‘strong’. We don’t even have to know how the function works, or whether it is consistent with round earth or whatnot. We just know that, given the height, there is an 80% chance of being within 2 miles of the observation, and a 70% chance of being within 1 mile.

Does that work better for you?

Re: Rowbotham experiment #9
« Reply #64 on: January 07, 2019, 03:15:53 AM »
Quote
Except that when using nautical miles it goes against your theory.

Using statute miles, in a nautical work, to provide evidence for your theory is deceptive and you should have provided both versions rather than choosing to deceive through misrepresentation.

If the measurements presented were miles then he should have used miles. If you want to know where some distance in nautical miles falls along the trendline simply convert that distance into miles and find the x coordinate whose output is that distance in miles you are using. Converting his use of miles to nautical miles would yield the same results as the actual distance is still the same and as long as the conversion is done correctly the equivalent distance in some other unit will by definition still be the same distance. If something is 1000m away and you want it converted to kilometers, that object is still 1000m away, only now your gragh has it as 1.000km. You are literally asking him to apply one additional coefficient which graphically appears as a compression or stretch unless a different scale is used to compensate; if the latter is true the two graphs will be identical

Offline edby

  • *
  • Posts: 1066
    • View Profile
Re: Rowbotham experiment #9
« Reply #65 on: January 07, 2019, 08:36:18 AM »
Quote
Except that when using nautical miles it goes against your theory.

Using statute miles, in a nautical work, to provide evidence for your theory is deceptive and you should have provided both versions rather than choosing to deceive through misrepresentation.

If the measurements presented were miles then he should have used miles. If you want to know where some distance in nautical miles falls along the trendline simply convert that distance into miles and find the x coordinate whose output is that distance in miles you are using. Converting his use of miles to nautical miles would yield the same results as the actual distance is still the same and as long as the conversion is done correctly the equivalent distance in some other unit will by definition still be the same distance. If something is 1000m away and you want it converted to kilometers, that object is still 1000m away, only now your gragh has it as 1.000km. You are literally asking him to apply one additional coefficient which graphically appears as a compression or stretch unless a different scale is used to compensate; if the latter is true the two graphs will be identical
I appreciate that, but it depends whether our function f(x) gives results in statute or nautical miles. We don't know exactly which units Findlay was using, and Rowbotham actually concedes this.

I did the calculation both ways and actually makes very little difference. That is because the correlation function is implicitly looking at deviations from average values.

*

Offline Pete Svarrior

  • e
  • Planar Moderator
  • *****
  • Posts: 8665
  • (◕‿◕✿)
    • View Profile
    • The Flat Earth Society
Re: Rowbotham experiment #9
« Reply #66 on: January 19, 2019, 01:42:54 PM »
Does that work better for you?
No. No amount of hand-waving will clear your logical fallacies and your attempt at overfitting a tiny subset of the data you're considering. You have to correct your errors, not restate them until I get bored.
Read the FAQ before asking your question - chances are we've already addressed it.
Follow the Flat Earth Society on Facebook and Twitter!


*mic stays stationary and earth accelerates upwards towards it*

Re: Rowbotham experiment #9
« Reply #67 on: January 19, 2019, 02:59:07 PM »
The comment about causation and correlation is another example of “heads I win, tails you lose” reasoning.
The globe model can predict how far one should be able to see objects from if we know their height above sea level. Now, in real life there may be refraction effects which make it harder to predict exactly but some correlation between the maximum distance an object can be seen and the prediction from the model would build confidence in the model.

Rowbotham casts doubt on the model by claiming that there are lots of sightings from much further than predicted by the model. The counter argument presented was that the non-UK data seems to be unreliable and from a different source to UK data and the UK data does actually correlate quite well with the globe earth model.
Then Pete swings in with “correlation doesn’t imply causation”. Now, he’s right but here we are testing a model which claims a correlation. The criticism is that the correlation doesn’t exist, ergo the model is wrong. If someone shows that a correlation does exist for a section of the data and that other data had a different, possibly less reliable source, then that can’t just be dismissed as meaningless
If you are making your claim without evidence then we can discard it without evidence.

*

Offline Pete Svarrior

  • e
  • Planar Moderator
  • *****
  • Posts: 8665
  • (◕‿◕✿)
    • View Profile
    • The Flat Earth Society
Re: Rowbotham experiment #9
« Reply #68 on: January 19, 2019, 03:09:47 PM »
then that can’t just be dismissed as meaningless
I agree - but unfortunately that's the situation we were presented with. I pointed out that this is likely a case of overfitting, and edby informed us that he likes England because he lives there. Any meaningful discussion died at that point.
Read the FAQ before asking your question - chances are we've already addressed it.
Follow the Flat Earth Society on Facebook and Twitter!


*mic stays stationary and earth accelerates upwards towards it*