31-05-2020 07:56 | |
James___★★★★★ (2854) |
duncan61 wrote: You are sounding like the Catholic church on this one. https://en.wikipedia.org/wiki/Galileo_thermometer Galileo was placed under house arrest for saying that the Earth orbited the Sun. This was until his death. In saying we cannot measure a temperature which Galileo also did is to support the Catholic church in saying that we are ignorant. I think the Catholic church accepts that it was wrong about Galileo. Edited on 31-05-2020 07:57 |
01-06-2020 03:34 | |
duncan61★★☆☆☆ (388) |
I have a Milwaukee temperature gun and the variance in any room in the house can be 10.C or more so I claim the average temperature of the room can not be known exactly so how the average global temperature can be known is also a mystery then the claim of something we do not know is slightly different from something we do not know makes interesting reading.Not sure about the Catholic Analogy |
01-06-2020 07:59 | |
IBdaMann★★★★★ (7086) |
duncan61 wrote: Regarding the temperature of a room, you notice that any single measurement can lead to an estimate that is off by perhaps 10 deg C. So every time you need to know the temperature of a room, you ask yourself "Will I be satisfied with a margin of error of 10 deg C?" If you are fine with that margin of error then you take one measurement and you are done. You record your temperature as X +/- 10 deg C. If that is not acceptable, you can measure the temperature in the eight corners of the room (presuming a rectangular prism shape) and the center, take a straight average and you should (depending on the room your mileage may vary) be operating within a margin of error of less than 2 deg C however you could fine-tune your estimate by statistically weighting your measurements using standard deviation. That will likely get you within one degree C margin of error. Again, of course everything depends on the specifics of the room. . A Spaghetti strainer with the faucet running, retains water- tmiddles Clouds don't trap heat. Clouds block cold. - Spongy Iris Printing dollars to pay debt doesn't increase the number of dollars. - keepit If Venus were a black body it would have a much much lower temperature than what we found there.- tmiddles Ah the "Valid Data" myth of ITN/IBD. - tmiddles Ceist - I couldn't agree with you more. But when money and religion are involved, and there are people who value them above all else, then the lies begin. - trafn You are completely misunderstanding their use of the word "accumulation"! - Climate Scientist. The Stefan-Boltzman equation doesn't come up with the correct temperature if greenhouse gases are not considered - Hank :*sigh* Not the "raw data" crap. - Leafsdude IB STILL hasn't explained what Planck's Law means. Just more hand waving that it applies to everything and more asserting that the greenhouse effect 'violates' it.- Ceist |
01-06-2020 16:32 | |
duncan61★★☆☆☆ (388) |
I get that IBDM and mentioned in a an earlier post on this thread I could do it with a pencil and piece of paper.But could I declare it is .8 warmer than 100 years ago with any certainty because I broke wind |
02-06-2020 02:34 | |
IBdaMann★★★★★ (7086) |
duncan61 wrote: I get that IBDM and mentioned in a an earlier post on this thread I could do it with a pencil and piece of paper.But could I declare it is .8 warmer than 100 years ago with any certainty because I broke wind No, you would not be able to say that with any sort of validity. As you and Harvey have noted, we cannot travel time into the past and verify anything. Speculation of the past does not carry very small margins of error; it carries very small margins of accuracy. The further back in time the speculation, the thinner the sliver of accuracy becomes. A Spaghetti strainer with the faucet running, retains water- tmiddles Clouds don't trap heat. Clouds block cold. - Spongy Iris Printing dollars to pay debt doesn't increase the number of dollars. - keepit If Venus were a black body it would have a much much lower temperature than what we found there.- tmiddles Ah the "Valid Data" myth of ITN/IBD. - tmiddles Ceist - I couldn't agree with you more. But when money and religion are involved, and there are people who value them above all else, then the lies begin. - trafn You are completely misunderstanding their use of the word "accumulation"! - Climate Scientist. The Stefan-Boltzman equation doesn't come up with the correct temperature if greenhouse gases are not considered - Hank :*sigh* Not the "raw data" crap. - Leafsdude IB STILL hasn't explained what Planck's Law means. Just more hand waving that it applies to everything and more asserting that the greenhouse effect 'violates' it.- Ceist |
02-06-2020 03:39 | |
James___★★★★★ (2854) |
IBdaMann wrote:duncan61 wrote: I get that IBDM and mentioned in a an earlier post on this thread I could do it with a pencil and piece of paper.But could I declare it is .8 warmer than 100 years ago with any certainty because I broke wind Any chance you and Harvey are involved in an intimate relationship? It's cool if you are but you shouldn't let that cloud your judgment. |
03-06-2020 13:46 | |
tmiddles★★★★★ (3276) |
duncan61 wrote: Well you're a bit cute there with manufacturing a question I never asked "measured for everywhere". I was clear that we were asking if you "had a clue". You seem intent on dodging that. Of course it's insane to say you cannot have a clue what the temperature inside your home is but are saying that? Because no you didn't. You said something about "measured for everywhere". duncan61 wrote:...I claim the average temperature of the room can not be known exactly...I don't believe temperature can be known exactly under any circumstance. If you want to make fake conditions go ahead. How about it all has to be notarized by GOD or we aren't allowed to discuss it! Reply to the post you want me to respond to and I will. I have never posted anything related to a change in sea level that I can recall. I think you're mistaken and it was a photo of melted ice above sea level. IBdaMann wrote:...you can measure the temperature in the eight corners of the room... standard deviation....get you within one degreeYep, sounds about right to me. gfm7175 wrote:WTF are you talking about? You see that here?: Into the Night wrote: |
03-06-2020 16:43 | |
gfm7175★★★☆☆ (941) |
tmiddles' homework, as posted earlier in this thread by ITN: The highest temperature recorded on Earth's surface was 134.1 degF. The lowest temperature recorded on Earth's surface was -128.6 degF. These measurements form the limits of variance, not the margin of error. It is not unusual for temperature to vary as much as 20 deg F per mile. Such times can take place across weather fronts, within range of mountain compression waves, or even the difference between a hot parking lot and a cool forest nearby. I use this value for the variance itself. With a single thermometer, therefore, using these numbers and limits of variance, it's accuracy of measurement is +-131 deg F. Using 10000 thermometers gives the same margin of error. We simply do not have enough thermometers in the world to reduce the margin of error below this value. |
04-06-2020 10:38 | |
tmiddles★★★★★ (3276) |
gfm7175 wrote: Ah I see. ITN was a bit sloppy and had that within a quote box attributed to me so I didn't notice it. I will show this as being ITN's comment: Into the Night wrote:....using these numbers and limits of variance, it's accuracy of measurement is +-131 deg F.So your margin of error is the difference between the highest and lowest temperatures recorded? Isn't that called the "variance" and not the "margin of error"? Also isn't that a bit arbitrary? Surely it's been hotter or colder than those measurements and no one happened to be there to record it. Edited on 04-06-2020 10:42 |
04-06-2020 20:54 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:gfm7175 wrote: RQAA. The Parrot Killer |
04-06-2020 22:35 | |
tmiddles★★★★★ (3276) |
Into the Night wrote:So you made a mistake and called the "variance" the "margin of error"? You most certainly never addressed your own post above before.tmiddles wrote:...Isn't that called the "variance" and not the "margin of error"? ... |
04-06-2020 22:43 | |
gfm7175★★★☆☆ (941) |
tmiddles wrote:Into the Night wrote:So you made a mistake and called the "variance" the "margin of error"? You most certainly never addressed your own post above before.tmiddles wrote:...Isn't that called the "variance" and not the "margin of error"? ... Reread what ITN said... ITN said that the "margin of error" was +-131 deg F. 134.1 + (-128.6) = 262.7 (aka "the difference between the highest and lowest temperatures recorded") 131 ≠ 262.7 Edited on 04-06-2020 22:49 |
04-06-2020 22:59 | |
tmiddles★★★★★ (3276) |
gfm7175 wrote:Yes which is clearly wrong if one means to identify the "margin of error" which is universally expressed as a positive number only and a percentage, not a unit value. That is if someone is talking about statistics and not some personal and private discipline. "A margin of error tells you how many percentage points your results will differ from the real population value. For example, a 95% confidence interval with a 4 percent margin of error means that your statistic will be within 4 percentage points of the real population value 95% of the time." Variance on the other hand is what ITN DID identify was the range of historical measurements. A fairly arbitrary and useless thing to identify. It remains entirely unclear, as he refuses to discuss what he wrote, what he meant. |
05-06-2020 02:49 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:Into the Night wrote:So you made a mistake and called the "variance" the "margin of error"? You most certainly never addressed your own post above before.tmiddles wrote:...Isn't that called the "variance" and not the "margin of error"? ... Don't put words in people's mouths. RQAA. The Parrot Killer |
05-06-2020 02:53 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:gfm7175 wrote:Yes which is clearly wrong if one means to identify the "margin of error" which is universally expressed as a positive number only and a percentage, not a unit value. Margin of error is not a percentage. tmiddles wrote: You deny statistical mathematics. You don't get to change it. tmiddles wrote: Not margin of error. There is no such thing as a 'confidence interval'. There is no 'interval' in statistical mathematics. tmiddles wrote: Observation limits are not variance. tmiddles wrote: Already discussed. RQAA. The Parrot Killer |
05-06-2020 14:32 | |
tmiddles★★★★★ (3276) |
Into the Night wrote:Did you learn statistics in school ITN? Would you kindly share a textbook that was used? I'm fascinated that you have this unique form of mathematics found nowhere else. |
05-06-2020 19:17 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:Into the Night wrote:Did you learn statistics in school ITN? Would you kindly share a textbook that was used? I'm fascinated that you have this unique form of mathematics found nowhere else. RQAA. The Parrot Killer |
05-06-2020 21:20 | |
tmiddles★★★★★ (3276) |
Into the Night wrote:tmiddles wrote:...kindly share a textbook that was used?... ITN what do you think of this one?: Introductory Statistics ISBN-10: 1938168208 Publish Date: Sep 19, 2013 Is it a good one? Has it been corrupted by Warmazombies or otherwise rendered itself random numbers or something? |
05-06-2020 21:51 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:Into the Night wrote:tmiddles wrote:...kindly share a textbook that was used?... RQAA. The Parrot Killer |
05-06-2020 22:24 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:Into the Night wrote:tmiddles wrote:Into the Night wrote:tmiddles wrote:...kindly share a textbook that was used?... RQAA. The Parrot Killer |
05-06-2020 22:38 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:Into the Night wrote:tmiddles wrote:Into the Night wrote:tmiddles wrote:...kindly share a textbook that was used?... RQAA. The Parrot Killer |
05-06-2020 22:59 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:Into the Night wrote:tmiddles wrote:Into the Night wrote:tmiddles wrote:...kindly share a textbook that was used?... Spam. RQAA. The Parrot Killer |
05-06-2020 23:01 | |
tmiddles★★★★★ (3276) |
Into the Night wrote:tmiddles wrote:Into the Night wrote:tmiddles wrote:...kindly share a textbook that was used?... And it says: "A confidence interval is another type of estimate but, instead of being just one number, it is an interval of numbers. It provides a range of reasonable values in which we expect the population parameter to fall. There is no guarantee that a given confidence interval does capture the parameter, but there is a predictable probability of success." But you said there was no such thing as a confidence interval in Statistics? Into the Night wrote:There is no such thing as a 'confidence interval'. There is no 'interval' in statistical mathematics.Are they warmazombies ITN? Because the textbook doesn't deal with Global Warming at all. |
05-06-2020 23:48 | |
IBdaMann★★★★★ (7086) |
tmiddles wrote: But you said there was no such thing as a confidence interval in Statistics? Into the Night is saying the same thing. You are mixing the words. Notice that the author writes ""It provides a range of reasonable values in which we expect the population parameter to fall." Those individual values must be specific values and cannot be "ranges" as you are asserting. For example, if I want to take the temperature of a room right now, I don't consider "confidence" ... I consider margin of error. However, if I am building a system that is going to heat that room then yes, I need to speak in terms of my confidence that the system will achieve a temperature within a particular margin of error ... and I will certainly have a range of conditions under which I am more confident than under others. In these cases, the margin of error is assumed as part of "the temperature" and I compute the probabilities that my goals will be met which will form a range of values. As you read that section, notice that the context is in predicting a future event, and in your confidence of certain events unfolding in a certain way ... vs the actual measurement of those events which is simply a here-and-now measurement with margin of error. Margin of Error: Measurement Confidence Level: Prediction of Future Events/Outcomes So, when Into the Night says that there is no confidence interval in measurement, he is correct. Measurement is not the prediction of events. Why? Because statistics is math, not science. Science, through math, predicts nature. Math alone can only give you probabilities. . A Spaghetti strainer with the faucet running, retains water- tmiddles Clouds don't trap heat. Clouds block cold. - Spongy Iris Printing dollars to pay debt doesn't increase the number of dollars. - keepit If Venus were a black body it would have a much much lower temperature than what we found there.- tmiddles Ah the "Valid Data" myth of ITN/IBD. - tmiddles Ceist - I couldn't agree with you more. But when money and religion are involved, and there are people who value them above all else, then the lies begin. - trafn You are completely misunderstanding their use of the word "accumulation"! - Climate Scientist. The Stefan-Boltzman equation doesn't come up with the correct temperature if greenhouse gases are not considered - Hank :*sigh* Not the "raw data" crap. - Leafsdude IB STILL hasn't explained what Planck's Law means. Just more hand waving that it applies to everything and more asserting that the greenhouse effect 'violates' it.- Ceist Edited on 05-06-2020 23:51 |
06-06-2020 02:32 | |
tmiddles★★★★★ (3276) |
IBdaMann wrote:I am mixing the words? I simply quoted ITN, here: Into the Night wrote:There is no such thing as a 'confidence interval'. There is no 'interval' in statistical mathematics.and then the textbook here: Introductory Statistics ISBN-10: 1938168208:A confidence interval is ...So while you might say that ITN is mixed up and doesn't realize that he agrees with the textbook you certainly can't mean that I have somehow manipulated what he had to say can you? IBdaMann wrote:...must be specific values and cannot be "ranges" as you are asserting.I was quoting the textbook IBD. I'd be very interested to know if you think they got it wrong. Maybe there is a better choice out there?: Introductory Statistics ISBN-10: 1938168208:A confidence interval is another type of estimate but, instead of being just one number, it is an interval of numbers. It provides a range of reasonable values in which we expect the population parameter to fall. ...So don't go giving me credit for the writings of Illowsky and Dean there in the text book. Are you saying they got it wrong? Could you recommend a better elementary text on Statistics? IBdaMann wrote:Confidence Level: Prediction of Future Events/OutcomesSo you're saying a confidence level only applies to future events and not the present? I don't think that's right. I think they are one in the same: The confidence interval is the estimate ± the margin of error.link The the "confidence" is usually 95% or 99% and never 100%. IBdaMann wrote:...Measurement is not the prediction of events.It is still an estimate which is no different really. A poll is probably the most frequently referenced activity with a margin of error and it is an estimate of the present. This all seems like an elaborate attempt to justify an error made by ITN. Why can't you guys just admit you mess up? everybody does. I don't think it matters one bit what is called science, math or religion. Things can simply be described as what they are and the labels aren't helpful. Edited on 06-06-2020 02:35 |
07-06-2020 00:47 | |
IBdaMann★★★★★ (7086) |
tmiddles wrote:I am mixing the words? I simply quoted ITN, here: Don't do that. You do the same thing with Trump, i.e. you take an exact quote out of context and then claim that the context doesn't matter because "look, this is exactly what he said/wrote." What Into the Night wrote was correct for the reasons I took the time to explain. tmiddles wrote:IBdaMann wrote:...must be specific values and cannot be "ranges" as you are asserting.I was quoting the textbook IBD. You quoted out of context and you misapplied the terms. Of course you and I can discuss our varying confidence levels of whatever and call it our range of confidence levels. Of course it won't be within the context used by the author of the book. When Into the Night specified that there is no range of confidence levels in a measurement, he was completely correct. Your insistence that context doesn't matter is dishonest and Into the Night isn't going to simply repeat an answer that he already gave. I'll repeat what I wrote previously: For measurements -> we discuss margin of error For predicting events (or results or findings) -> we discuss confidence levels ... *and* ... confidence levels pertain to measurement values that include margin of error. tmiddles wrote:IBdaMann wrote:Confidence Level: Prediction of Future Events/OutcomesSo you're saying a confidence level only applies to future events and not the present? I don't think that's right. I don't think you are clear on what you are thinking. If you measure a wall and find it to be of length 27 meters 13.443 cm +/- 0.09 cm ... your tolerance is the +/- 0.09 cm. Confidence level does not apply because you have measured the wall and you are now 100% confident, i.e. absolutely certain, that the wall is 27 meters 13.443 cm +/- 0.09 cm. If before you measure the length of the wall you were to study the construction processes of the builder and find that this type of wall is supposed to be built to within 0.7 cm of the specified length which happens to be 27 meters 11 cm in this case, and you find that the builder normally overbuilds because of reliance on uncalibrated equipment, you can declare a low confidence that the wall will meet specifications (because you haven't measured yet). You can base your confidence levels on the previous measurements of the varying types of walls (i.e. data) and develop a range of confidence levels for different types of walls and for the differing uncalibrated equipment that was used. None of this pertains to, or affects, the actual measurements of any walls. Feel free to work through the past, present and future tense as necessary to get a better understanding of the full context. Perhaps one way for you to look at it is to consider that "confidence" only applies in a situation where we do not know and must speculate based on past data and experience, whereas margin of error applies to measurements where we are "finding out" and getting an answer. They are not the same thing. . A Spaghetti strainer with the faucet running, retains water- tmiddles Clouds don't trap heat. Clouds block cold. - Spongy Iris Printing dollars to pay debt doesn't increase the number of dollars. - keepit If Venus were a black body it would have a much much lower temperature than what we found there.- tmiddles Ah the "Valid Data" myth of ITN/IBD. - tmiddles Ceist - I couldn't agree with you more. But when money and religion are involved, and there are people who value them above all else, then the lies begin. - trafn You are completely misunderstanding their use of the word "accumulation"! - Climate Scientist. The Stefan-Boltzman equation doesn't come up with the correct temperature if greenhouse gases are not considered - Hank :*sigh* Not the "raw data" crap. - Leafsdude IB STILL hasn't explained what Planck's Law means. Just more hand waving that it applies to everything and more asserting that the greenhouse effect 'violates' it.- Ceist |
07-06-2020 02:20 | |
tmiddles★★★★★ (3276) |
IBdaMann wrote:I can write everytime "I fail to see how there is context to change the meaning here" if you like but I assume it is implied. I was correcting your error is stating: IBdaMann wrote:You are mixing the words....."ranges" as you are asserting.As I am simply, and without creativity, quoting where the words are introduced in the text book. I am asserting that it is in the textbook. IBdaMann wrote:No you failed to explain how ITN was correct in asserting that a "confidence interval" is a nonexistent thing in statistics, which is what he said: Into the Night wrote: There is no such thing as a 'confidence interval'. There is no 'interval' in statistical mathematics.He didn't say sometimes. IBdaMann wrote:One more time: I simply quoted the textbook where the terms were introduced. I did so by way of a question. So you can pretend I have some "assertion" here but it's is only that the terms, which ITN denies, exist in the text book. Note he did not say they were defined differently, he said they did not exist for the subject matter. IBdaMann wrote:For measurements -> we discuss margin of errorAnd I will quote the citation I provided again and maybe you'll respond: I think they are one in the same: The confidence interval is the estimate ± the margin of error.link The the "confidence" is usually 95% or 99% and never 100%. IBdaMann wrote:...you have measured the wall and you are now 100% confident, i.e. absolutely certain, that the wall is 27 meters 13.443 cm +/- 0.09 cm.And where does the +/- 0.09 cm come from? Whay isn't it +/- 0.08? This is a calculation done in satistics and guess what? IT REQUIRES A CONFIDENCE LEVEL (usually 95% or 99% are chosen depending on the subject matter): How to calculate margin of errorlink Is someone is "absolutely certain" they nailed the margin of error they are mistaken. Once again you and ITN stand on rare ground in opposition to EVERY TEXTBOOK ON STATISTICS THAT HAS EVER BEEN PUBLISHED and that is elementary statistics. Got nothing to do with AGW or warmazombies or any excuse you guys might have for how the textbooks are corrupted. Edited on 07-06-2020 02:23 |
07-06-2020 06:08 | |
IBdaMann★★★★★ (7086) |
tmiddles wrote: As I am simply, and without creativity, quoting where the words are introduced in the text book. I am asserting that it is in the textbook. I'm not going to belabor this point. I followed your conversation and I understood what Into the Night meant. Either you want to understand the concepts (in which case Into the Night is correct about not conflating confidence with margin of error) or you want to quibble about what words have been written independent of context, then sure, you are correct. This point does not interest me any further. tmiddles wrote: I think they are one in the same: The confidence interval is the estimate ± the margin of error. Incorrect. Look, this is something so stupid that I would otherwise be inclined to rattle of a few sentences to that effect, but instead I will just ask you whether you can see the complete difference between the varying confidence levels that different people hold that Khabib Nurmagomedov will win his next UFC fight ... and the actual results of his next fight? Confidence levels <- vs. -> Actual Results They are not the same thing. One is speculation. One is a measure. Confidence levels never become data. Insisting that they are one and the same is ... well, stupid. tmiddles wrote: And where does the +/- 0.09 cm come from? The tolerance of the measuring equipment, as I wrote in my post. It's not margin of error. It is equipment tolerance. tmiddles wrote:Once again you and ITN stand on rare ground in opposition to EVERY TEXTBOOK ON STATISTICS THAT HAS EVER BEEN PUBLISHED You can't even express my position, much less speak for every statistic book ever published, despite the bold type. [ D I S M I S S E D ] . A Spaghetti strainer with the faucet running, retains water- tmiddles Clouds don't trap heat. Clouds block cold. - Spongy Iris Printing dollars to pay debt doesn't increase the number of dollars. - keepit If Venus were a black body it would have a much much lower temperature than what we found there.- tmiddles Ah the "Valid Data" myth of ITN/IBD. - tmiddles Ceist - I couldn't agree with you more. But when money and religion are involved, and there are people who value them above all else, then the lies begin. - trafn You are completely misunderstanding their use of the word "accumulation"! - Climate Scientist. The Stefan-Boltzman equation doesn't come up with the correct temperature if greenhouse gases are not considered - Hank :*sigh* Not the "raw data" crap. - Leafsdude IB STILL hasn't explained what Planck's Law means. Just more hand waving that it applies to everything and more asserting that the greenhouse effect 'violates' it.- Ceist |
07-06-2020 06:23 | |
duncan61★★☆☆☆ (388) |
You 2 should move in together then we can remake the series The odd couple |
07-06-2020 09:29 | |
tmiddles★★★★★ (3276) |
IBdaMann wrote:...Into the Night is correct about not conflating confidence with margin of error...Well there is not and never has been a margin of error that didn't have a confidence level so he is wrong no matter how you want to look at it. There is no 100% confidence level so it is always part of a margin of error. Every textbook on statistics, and every use of statistics, will confirm this. IBdaMann wrote:I will just ask you ...varying confidence levels that different people hold that Khabib Nurmagomedov will win his next UFC fight...You are confusing the emotion of confidence with the "confidence level" which is ALWAYS ALWAYS ALWAYS part of any use of statitstics. It is not possible to use statistics without including a confidence level, which is not an emotional state but a probability. You can have a 95% confidence level that the margin of error is +/- 0.9 and so on. You can NEVER be "certain" and disregard that confidence/probability dimension to statistics. You and ITN would seem to be entirely ignorant on the subject, which is frankly surprising given all your boasting up till now. tmiddles wrote:You conveniently ignored part of my post above. So why don't you show a margin of error being calculated for something? Anything at all. I'll be very interested to see how you avoid a confidence level. So here we go. Examples from Statistics texts on confidence interval: Introductory Statistics ISBN-10: 1938168208 Publish Date: Sep 19, 2013 "A confidence interval is another type of estimate but, instead of being just one number, it is an interval of numbers. It provides a range of reasonable values in which we expect the population parameter to fall. There is no guarantee that a given confidence interval does capture the parameter, but there is a predictable probability of success." Introductory Statistics ISBN 13: 9781453344873 Publish Date: Nov 29, 2017 "A problem with a point estimate is that it gives no indication of how reliable the estimate is. In contrast, in this chapter we learn about interval estimation. In brief, in the case of estimating a population mean μ we use a formula to compute from the data a number E, called the margin of error of the estimate, and form the interval [x−−−E,x−−+E]. We do this in such a way that a certain proportion, say 95%, of all the intervals constructed from sample data by means of this formula contain the unknown parameter μ. Such an interval is called a 95% confidence interval for μ." Think Stats ISBN 13: 9781491907337 Publish Date: 2014 "13.7 Confidence intervals Kaplan-Meier analysis yields a single estimate of the survival curve, but it is also important to quantify the uncertainty of the estimate. As usual, there are three possible sources of error: measurement error, sampling error, and modeling error." OpenIntro Statistics Third Edition Publish Date: 2015 "4.2 CONFIDENCE INTERVALS A point estimate provides a single plausible value for a parameter. However, a point estimate is rarely perfect; usually there is some error in the estimate. Instead of supplying just a point estimate of a parameter, a next logical step would be to provide a plausible range of values for the parameter....A point estimate provides a single plausible value for a parameter. However, a point estimate is rarely perfect; usually there is some error in the estimate. Instead of supplying just a point estimate of a parameter, a next logical step would be to provide a plausible range of values for the parameter." Every single instruction or employment of statistics will have a confidence level. To claim to understand statistics but to deny this is truly amazing. THIS IS HIGH SCHOOL MATH |
07-06-2020 11:27 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:...deleted 25c...25c...10 (margin of error<->confidence)...30...4b...4d...34...25c...25c...22 (confidence level)...25...25n...25c...25n..12...35c...denial of mathematics...30...17...29...29...22 (confidence level)...4a...4b...22 (confidence level)...25n...25n...25c...25h...25g...25g...word salad...4b...4a...25c1...4b...4a...25g...25c..25h....word salad...25c...22 (confidence level)...35c...12...39f...39j...39m... No argument presented. Word salad. Denial of mathematics (statistical and probability). Denial of randR. Math errors. False authority fallacies. Buzzword fallacies. RQAA. The Parrot Killer |
07-06-2020 13:13 | |
tmiddles★★★★★ (3276) |
Into the Night wrote:Well this stands as proof your understanding of statistics is nonexistent, as is IBDs. |
08-06-2020 01:00 | |
IBdaMann★★★★★ (7086) |
tmiddles wrote:THIS IS HIGH SCHOOL MATH And it is beyond you. I can see why you wander aimlessly. This is what doubling down on stupid looks like. tmiddles wrote: You are confusing the emotion of confidence with the "confidence level" which is ALWAYS ALWAYS ALWAYS part of any use of statitstics. Nope. Just two days ago I performed a least-squares analysis and there was no question of confidence level. None. Your stupid absolutes serve as great doubling down material. tmiddles wrote: Well there is not and never has been a margin of error that didn't have a confidence level The exact opposite. There is no such thing as any margin of error with any sort of confidence level. You have to switch that around. There is no confidence level that doesn't have a margin of error as a factor in its basis. tmiddles wrote: ... so he is wrong no matter how you want to look at it. ... so you are wrong no matter how you want to look at it. tmiddles wrote: There is no 100% confidence level Yes there is ... it just depends on the context. I gave you an example of 100% confidence but it was within the context of measurement. Here's another. If you are at the black jack table in a Vegas casino, there is no uncertainty about you having a nine and a four, i.e. that you are holding thirteen. Now if you completely change the context from the measure of what you have to an assessment of your standing, to an analysis of what you should do, etc. yes, under an entirely different context you can have varying confidence levels, none of which will involve certainty. So, repeating for the third time, just in yet another different way: 1) There is certainty in what you have once you measure it. 2) There is no certainty in predicting future results/findings and thus you perform statistical methods to arrive at varying confidence levels for courses of action. How about I leave it up to you to figure this out and I don't repeat it. This is growing very tiresome. tmiddles wrote: Every textbook on statistics, and every use of statistics, will confirm this. Every statistics textbook should confirm what I just wrote above. Note: You just tried to pretend to speak for all textbooks on statistics. I should very well stop reading your post. tmiddles wrote: It is not possible to use statistics without including a confidence level, which is not an emotional state but a probability. Go learn some statistics. You are gibbering. Yes, "confidence level" is absolutely tied to the subjective human emotion of confidence. You are a Bozo for thinking otherwise ... actually, you are just a mathematically incompetent Bozo spouting off without having a clue. Statisitical methods are used by humans to help them make decisions based on data, so they can be more confident in the decisions they make. Say it with me ... "emotion of confidence." Did it feel good? Let's do it again ... "emotion of confidence." Yes, that is exactly what it is. So now it's your turn to get back to my question and give me an answer: Are you smart enough to discern the semantic chasm separating a human confidence level in whether Khabib Nurmagomedov will win his next fight and the actual results of the fight? If you are able to discern the difference then everything you wrote up in your latest post was an intentional waste of time. If you are not smart enough to discern the difference then you need a lot of help on a more fundamental level. How shall we proceed? tmiddles wrote: You and ITN would seem to be entirely ignorant on the subject, I will keep that under advisement, especially that it is coming from you. tmiddles wrote: ... which is frankly surprising given all your boasting up till now. Now you have piqued my interest. Where have I so boasted? In the meantime, let's bore ourselves to tears addressing your additional babblings: tmiddles wrote: "A confidence interval is another type of estimate but, instead of being just one number, it is an interval of numbers. ... but is not a measure. It uses measurements and specific data values "to provide a range of reasonable values in which we expect the population parameter to fall." Like computing the global average temperature when you don't have a valid dataset that can get you within your desired margin of error, so you resort to different confidence levels in different values, but you really don't know. tmiddles wrote: "A problem with a point estimate is that it gives no indication of how reliable the estimate is. Also not a measure ... the author is talking about an estimate. If you read further the author appears to be leading into the topic of standard deviation and in simply generating future probabilities based on previous results. The "95% rule" means two sigmas. Your droning on and on about there being no absolute certainty comes from the "99.7% rule" which means three sigmas. You're not hitting on anything new. The whole idea of "Six Sigma" is to get manufacturing tolerance down to 99.9999% defect-free, acknowledging that there can never be 100% elimination of errors/defects. Yawn. tmiddles wrote: As usual, there are three possible sources of error: measurement error, sampling error, and modeling error." Yep, I covered the first one. We can determine the tolerance of measuring equipment so we include that in all measurements. Sampling error comes from not determining the target margin of error in advance and thus sampling too few, or in the wrong locations, wrong amounts, at the wrong times, etc... Modeling error comes from incorporating false assumptions into your model, such as inserting predetermined conclusions. That will certainly get your model tossed. tmiddles wrote: However, a point estimate is rarely perfect; All of thes examples that you have cut-n-pasted are for estimates and not for measurements. This is what happens when you copy-paste blind without having read the material, but instead simply did a word search and posted snippets out of context. . Attached image: Edited on 08-06-2020 01:11 |
08-06-2020 03:42 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:Into the Night wrote:Well this stands as proof your understanding of statistics is nonexistent, as is IBDs. Fallacies are not a proof. Mantra 8...39n The Parrot Killer |
08-06-2020 04:57 | |
tmiddles★★★★★ (3276) |
IBdaMann wrote:...a least-squares analysis and there was no question of confidence level.Because you didn't have a margin of error. A confidence level is ALWAY associated with ANY margin of error PERIOD. I can perform addition and subtraction without a confidence level too! IBdaMann wrote:You mean this right?:tmiddles wrote: There is no 100% confidence level...I gave you an example of 100% confidence but it was within the context of measurement. IBdaMann wrote:If you measure a wall and find it to be of length 27 meters 13.443 cm +/- 0.09 cm ... your tolerance is the +/- 0.09 cm. Confidence level does not apply because you have measured the wall and you are now 100% confident, i.e. absolutely certain, that the wall is 27 meters 13.443 cm +/- 0.09 cm....The tolerance of the measuring equipment, as I wrote in my post. It's not margin of error. It is equipment tolerance. Guess what? The tolerance of the measuring equipment has a confidence level! Just because a manufacturer does the stats for you doesn't mean they don't exist. Read up: METROLOGY AND CALIBRATION Certified Quality Engineer exam preparation material and information. Confidence Level: The probability that a measured value will be inside of the tolerance limits. The confidence level for standards is usually set at 95%. IBdaMann wrote:... black jack table ...no uncertainty about you having a nine and a four,Do you mean there is no uncertainty that your eyes are seeing something in front of you that is not a measurement, estimate or anything that relates to statistics even remotely? That you are able to name the cards because you can read them? How about an example relevant to statistics. The measurement was. IBdaMann wrote:No there never is 100% certainty. There will never be an instrument capable of that. You are dead wrong. IBdaMann wrote: i.e. that you are holding thirteen.That we name the cards with numbers doesn't change that your example is entirely based on abstract human language and has nothing to do with measurement/estimation. It is the same as holding two magic the gathering cards and identifying them by name. IBdaMann wrote:...an analysis of what you should do,...you can have varying confidence levels,Again "Confidence Interval/Level" has nothing at all to do with human emotion, judgement or anything other than probability. IBdaMann wrote:Every statistics textbook should confirm what I just wrote above.None do because it's dead wrong, made up, garbage. I won't even ask for a citation because none exist. A measure is always an estimate. But let's take an example: Estimating Measurement Uncertainty "Evaluate the Uncertainty Due to the Calibration Standard and/or Instrumentation The next step is to review the calibration data from the calibration certificate for the calibration standard or test instrumentation that will be used for the measurement. Typically, when uncertainty is stated directly on calibration certificates it will be the expanded uncertainty with a coverage factor of two and a confidence level of 95%. " A tolerance IS a margin of error WITH a confidence level. But let's go to the real gold standard: https://en.wikipedia.org/wiki/Tolerance_interval "A tolerance interval is a statistical interval within which, with some confidence level, " let me guess, you allege warmazombie corruption? Into the Night wrote:All four textbooks ITN? |
08-06-2020 06:22 | |
IBdaMann★★★★★ (7086) |
tmiddles wrote: A confidence level is ALWAY associated with ANY margin of error PERIOD. Correct. You just made my point. Thank you. A confidence level is based on margin of error which includes factos like tolerance ... not the reverse. tmiddles wrote: Guess what? The tolerance of the measuring equipment has a confidence level! Yes, of course, but that is a completely separate uncertainty. Heisenberg taught us that there is uncertainty in everything, but that doesn't make all uncertainties the same. If you wish to get an accurate global average temperature for which you have a 95% confidence that it is within a 5dC margin of error then that will be a different confidence than any manufacturer's confidence in any measuring device. Your confidence level in your final calculation will be based on your methodology ... which will have its own margin of error. The margin of error does not have a confidence level, the methodology does. If you then choose a different methodology you will have a different margin of error and a different confidence in the end result. tmiddles wrote: Do you mean there is no uncertainty that your eyes are seeing something in front of you that is not a measurement, estimate or anything that relates to statistics even remotely? It is the result of a random draw. It is the "data" just as if you were rolling dice at the craps table. ... or are you under the impression that I cannot perform a statistical analysis on two people playing black jack to see who resulted in a statistically better performance? So, yes, measurement, die rolls, black jack hands, and any other realized results, data and observations are the domain of statistical analysis. So, your dodge is dismissed and you are mistaken. You also insist on EVADING my question about Khabib Nurmagomedov so I'm out until you do. . That you are able to name the cards because you can read them? How about an example relevant to statistics. The measurement was. IBdaMann wrote:No there never is 100% certainty. There will never be an instrument capable of that. You are dead wrong. IBdaMann wrote: i.e. that you are holding thirteen.That we name the cards with numbers doesn't change that your example is entirely based on abstract human language and has nothing to do with measurement/estimation. It is the same as holding two magic the gathering cards and identifying them by name. IBdaMann wrote:...an analysis of what you should do,...you can have varying confidence levels,Again "Confidence Interval/Level" has nothing at all to do with human emotion, judgement or anything other than probability. IBdaMann wrote:Every statistics textbook should confirm what I just wrote above.None do because it's dead wrong, made up, garbage. I won't even ask for a citation because none exist. A measure is always an estimate. But let's take an example: Estimating Measurement Uncertainty "Evaluate the Uncertainty Due to the Calibration Standard and/or Instrumentation The next step is to review the calibration data from the calibration certificate for the calibration standard or test instrumentation that will be used for the measurement. Typically, when uncertainty is stated directly on calibration certificates it will be the expanded uncertainty with a coverage factor of two and a confidence level of 95%. " A tolerance IS a margin of error WITH a confidence level. But let's go to the real gold standard: https://en.wikipedia.org/wiki/Tolerance_interval "A tolerance interval is a statistical interval within which, with some confidence level, " let me guess, you allege warmazombie corruption? Into the Night wrote:All four textbooks ITN?[/quote] A Spaghetti strainer with the faucet running, retains water- tmiddles Clouds don't trap heat. Clouds block cold. - Spongy Iris Printing dollars to pay debt doesn't increase the number of dollars. - keepit If Venus were a black body it would have a much much lower temperature than what we found there.- tmiddles Ah the "Valid Data" myth of ITN/IBD. - tmiddles Ceist - I couldn't agree with you more. But when money and religion are involved, and there are people who value them above all else, then the lies begin. - trafn You are completely misunderstanding their use of the word "accumulation"! - Climate Scientist. The Stefan-Boltzman equation doesn't come up with the correct temperature if greenhouse gases are not considered - Hank :*sigh* Not the "raw data" crap. - Leafsdude IB STILL hasn't explained what Planck's Law means. Just more hand waving that it applies to everything and more asserting that the greenhouse effect 'violates' it.- Ceist |
08-06-2020 10:30 | |
tmiddles★★★★★ (3276) |
IBdaMann wrote:Uh back it up there. You stated: IBdaMann wrote: IBdaMann wrote:So you said there is no confidence level for the measurement of the wall's +/- 0.09 cm tolerance. Heisenberg can't help you with how dead wrong you remain about that.tmiddles wrote: And where does the +/- 0.09 cm come from?The tolerance of the measuring equipment, as I wrote in my post. It's not margin of error. It is equipment tolerance. ITN said: Into the Night wrote:There is no such thing as a 'confidence interval'. There is no 'interval' in statistical mathematics.and you said: IBdaMann wrote:...when Into the Night says that there is no confidence interval in measurement, he is correct.You are both dead wrong, every textbook on statistics contradicts you, every spec sheet for a measuring device contradicts you! Nothing you've said clears up or address what you've said above. IBdaMann wrote:You also insist on EVADING my question about Khabib Nurmagomedov so I'm out until you do.I told you and I'll tell you again that the "confidence level" in statistics has NOTHING to do with human thought, decision making, emotion or anything related to asking people what they think or psychological confidence. It is a statistics term meaning the probability a sought for result is within the margin or error. You cannot have a margin of error without a confidence level. I strongly suspect you're just pulling my leg with such a stupid assertion it's related to feeling confident. I've now answered that 3 times. If you're going to say I didn't you'll need to explain why. |
08-06-2020 14:48 | |
IBdaMann★★★★★ (7086) |
tmiddles wrote:IBdaMann wrote:You also insist on EVADING my question about Khabib Nurmagomedov so I'm out until you do.I told you and I'll tell you again that the "confidence level" in statistics has NOTHING to do with human thought, decision making, emotion or anything related to asking people what they think or psychological confidence. ... and I corrected your stupid assertion. It is precisely the human emotion of confidence for the purpose of making decisions. So I need a better answer. Do you recognize the difference between a confidence level of a Khabib Nurmagomedov victory (speculation) and whatever the results of the fight happen to be (the determination)? Once we have the results of the fight, there is no longer any question of confidence level. We have certainty. Just as when you receive your hand in black jack, you know what you have with 100% certainty. You may be counting cards and have a high confidence level about whether you will get a ten or a face card, which will guide your decision, but you won't know until you receive your card. You won't know the result of the fight until the fight happens, at which point "confidence" ceases to exist and certainty takes its place. By the way, when Khabib Nurmagomedov fights the next time, his gloves will be expected to weigh at least four ounces and they can measure before the fight ... but there is uncertainty with any measure. He might have 3.9998 oz gloves but the scale might indicate 4.01 oz. despite a high confidence level in the accuracy of the scale. None of this will affect the odds at the sportsbook. So, do you recognize the difference or not? . A Spaghetti strainer with the faucet running, retains water- tmiddles Clouds don't trap heat. Clouds block cold. - Spongy Iris Printing dollars to pay debt doesn't increase the number of dollars. - keepit If Venus were a black body it would have a much much lower temperature than what we found there.- tmiddles Ah the "Valid Data" myth of ITN/IBD. - tmiddles Ceist - I couldn't agree with you more. But when money and religion are involved, and there are people who value them above all else, then the lies begin. - trafn You are completely misunderstanding their use of the word "accumulation"! - Climate Scientist. The Stefan-Boltzman equation doesn't come up with the correct temperature if greenhouse gases are not considered - Hank :*sigh* Not the "raw data" crap. - Leafsdude IB STILL hasn't explained what Planck's Law means. Just more hand waving that it applies to everything and more asserting that the greenhouse effect 'violates' it.- Ceist |
08-06-2020 18:36 | |
gfm7175★★★☆☆ (941) |
I recognize the difference, and it's been painful having to see this explained numerous times. You pain me tmiddles. |
08-06-2020 21:05 | |
Into the Night★★★★★ (12800) |
tmiddles wrote:IBdaMann wrote:Uh back it up there. Heisenberg is not involved. There is no such thing as 'confidence level' or 'confidence interval' in statistical math. Margin of error is not related to anything like 'confidence level' or 'confidence interval'. You are simply making shit up. tmiddles wrote: Tolerance is not 'confidence' anything. It is simply the accuracy of the instrument (whether its a ruler or a thermometer). Is not margin of error. You keep trying to equivocate these two as the same as well. They are not. tmiddles wrote:IBdaMann wrote:You also insist on EVADING my question about Khabib Nurmagomedov so I'm out until you do.I told you and I'll tell you again that the "confidence level" in statistics has NOTHING to do with human thought, decision making, emotion or anything related to asking people what they think or psychological confidence. There is no such thing as 'confidence level' in statistical math. You are making shit up. tmiddles wrote: Lie. tmiddles wrote: Here you are denying probability math as well. There is no 'sought for result' in statistical math. The summary result is always within the margin of error...100% of the time. There is no 'should be' in math or science. tmiddles wrote: BS. tmiddles wrote: BS. tmiddles wrote: And it's the same BS each time. tmiddles wrote: No need. You just like to repeat your made up shit over and over, just like you repeat the same questions that have already been answered over and over. The Parrot Killer |
Threads | Replies | Last post |
GRACE satellites to be shut down | 1 | 16-09-2017 22:11 |