jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Jul 1, 2010 0:58:51 GMT
This graph says it all..all cycle 24 spots are included. I don't think so. E.g. on howard.astro.ucla.edu/pub/obs/drawings/dr100602.jpg there were six spots. You say all six were included? Or only the biggest and darkest one? above the threshold? Thus making a biased selection. You don't have an understanding of how the measurements are taken, all 6 were included. It is explained clearly on the website. Basically the whole region is measured including any specks that might be included for any region that exceeds 23 pixels in total. Only the peak measurements are recorded thus removing any bias. This is a true indication of the magnetic strength of SC24.
|
|
|
Post by lsvalgaard on Jul 1, 2010 5:26:52 GMT
You don't have an understanding of how the measurements are taken, all 6 were included. It is explained clearly on the website. Basically the whole region is measured including any specks that might be included for any region that exceeds 23 pixels in total. Only the peak measurements are recorded thus removing any bias. This is a true indication of the magnetic strength of SC24. "Only the peak measurements are recorded thus removing any bias"Well, that introduces the bias right there. F10.7 is the true indicator of SC24 [and any other cycle].
|
|
jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Jul 1, 2010 7:30:30 GMT
You don't have an understanding of how the measurements are taken, all 6 were included. It is explained clearly on the website. Basically the whole region is measured including any specks that might be included for any region that exceeds 23 pixels in total. Only the peak measurements are recorded thus removing any bias. This is a true indication of the magnetic strength of SC24. "Only the peak measurements are recorded thus removing any bias"Well, that introduces the bias right there. F10.7 is the true indicator of SC24 [and any other cycle]. No bias, all days are measured with the highest reading for that group making the record. Much better than random selection. I see todays reading at 82%, a new SC24 record.
|
|
|
Post by lsvalgaard on Jul 1, 2010 13:21:24 GMT
"Only the peak measurements are recorded thus removing any bias"Well, that introduces the bias right there. F10.7 is the true indicator of SC24 [and any other cycle]. No bias, all days are measured with the highest reading for that group making the record. Much better than random selection. I see todays reading at 82%, a new SC24 record. Every day I measure the height of the children on our local school bus and only record the tallest one. Thus I bias the record against the taller end of the distribution. Or it would be like proving global warming by only recording the highest temperature each year.
|
|
jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Jul 2, 2010 2:10:08 GMT
No bias, all days are measured with the highest reading for that group making the record. Much better than random selection. I see todays reading at 82%, a new SC24 record. Every day I measure the height of the children on our local school bus and only record the tallest one. Thus I bias the record against the taller end of the distribution. Or it would be like proving global warming by only recording the highest temperature each year. Your problem is that you do not measure every day. Everyday I measure the height of every child on the bus every 4 hours , these mythical children have the ability to shrink and grow. I record the height of each child at their tallest point on their tallest day and plot each child's results to obtain a growth trend. This is better than measuring once a month which quite often falls on a school holiday.
|
|
|
Post by lsvalgaard on Jul 2, 2010 5:27:51 GMT
Every day I measure the height of the children on our local school bus and only record the tallest one. Thus I bias the record against the taller end of the distribution. Or it would be like proving global warming by only recording the highest temperature each year. Your problem is that you do not measure every day. For an unbiased sample you do not need to measure every day, as long as your selection criteria does not depend on anything you measure. This simple statistical truth seems to escape you, yet is used in any and all polls.
|
|
jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Jul 2, 2010 6:31:56 GMT
Your problem is that you do not measure every day. For an unbiased sample you do not need to measure every day, as long as your selection criteria does not depend on anything you measure. This simple statistical truth seems to escape you, yet is used in any and all polls. What seems to be escaping you is that measuring everything is better than random selection. But we do need to remember things are different in your universe The L&P method will eventually show the same trend, but its bit mind numbing waiting for the obvious result.
|
|
|
Post by lsvalgaard on Jul 2, 2010 7:06:32 GMT
For an unbiased sample you do not need to measure every day, as long as your selection criteria does not depend on anything you measure. This simple statistical truth seems to escape you, yet is used in any and all polls. What seems to be escaping you is that measuring everything is better than random selection. But we do need to remember things are different in your universe The L&P method will eventually show the same trend, but its bit mind numbing waiting for the obvious result. Except you do not use everything as you only retain the highest value for each group. This is deliberate bias, but is understandable considering your reason for the bias.
|
|
jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Jul 2, 2010 8:08:12 GMT
What seems to be escaping you is that measuring everything is better than random selection. But we do need to remember things are different in your universe The L&P method will eventually show the same trend, but its bit mind numbing waiting for the obvious result. Except you do not use everything as you only retain the highest value for each group. This is deliberate bias, but is understandable considering your reason for the bias. By only recording the highest reading of each group you get a true measure of the solar potential for magnetic strength, nothing can be hidden or missed as in your prescribed random measure. Your insistence to refuse to look at the recorded facts is disturbing. Science is about facts not agenda. You are on a hiding to nothing on this one.
|
|
|
Post by sranders on Jul 2, 2010 8:18:45 GMT
Let's try another analogy.
If I want to find out how well my front yard grass is growing, do I:
a) measure 100 randomly selected grass blades every day that it is not raining, or
b) measure the 100 longest grass blades every day that it is not raining?
I think b) would be biased toward an indication of better grass growth, and a) would not be. In fact, the only appropriate approach to monitoring lawn health is a).
Scott
|
|
jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Jul 2, 2010 8:49:08 GMT
Let's try another analogy. If I want to find out how well my front yard grass is growing, do I: a) measure 100 randomly selected grass blades every day that it is not raining, or b) measure the 100 longest grass blades every day that it is not raining? I think b) would be biased toward an indication of better grass growth, and a) would not be. In fact, the only appropriate approach to monitoring lawn health is a). Scott your analogy is weak....measure every grass blade over every day and there is no bias, what we are trying to achieve is how high a grass blade can grow. The rain, sunshine, nutrient content etc (growth potential) is the same as the magnetic strength of the Sun. Its a bit like randomly selecting laps in a F1 qualifying session to determine who has the fastest car. Pole position is determined by the fastest car, by recording all laps this can be determined accurately.
|
|
|
Post by ncfcadam on Jul 2, 2010 9:12:52 GMT
By only recording the highest reading of each group you get a true measure of the solar potential for magnetic strength, nothing can be hidden or missed as in your prescribed random measure. Your insistence to refuse to look at the recorded facts is disturbing. Science is about facts not agenda. Your on a hiding to nothing on this one. He certainly is on a hiding to nothing: I don't think you'll ever understand what sample selection bias is. I'm frankly amazed at his patience.
|
|
jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Jul 2, 2010 9:16:49 GMT
By only recording the highest reading of each group you get a true measure of the solar potential for magnetic strength, nothing can be hidden or missed as in your prescribed random measure. Your insistence to refuse to look at the recorded facts is disturbing. Science is about facts not agenda. Your on a hiding to nothing on this one. He certainly is on a hiding to nothing: I don't think you'll ever understand what sample selection bias is. I'm frankly amazed at his patience. And that seems to be your only argument. Random selection is better than measuring all. Very poor.
|
|
|
Post by ncfcadam on Jul 2, 2010 9:39:34 GMT
He certainly is on a hiding to nothing: I don't think you'll ever understand what sample selection bias is. I'm frankly amazed at his patience. And that seems to be your only argument. Random selection is better than measuring all. Very poor. Strawman. Nobody has said that random selection is better than measuring everything. It's physically impossible to measure all spots anyway because we can't see those on the side of the sun facing away from us. However, random selection is better than biased selection.
|
|
jinki
Level 3 Rank
Posts: 123
|
Post by jinki on Jul 2, 2010 9:45:13 GMT
And that seems to be your only argument. Random selection is better than measuring all. Very poor. Strawman. Nobody has said that random selection is better than measuring everything. It's physically impossible to measure all spots anyway because we can't see those on the side of the sun facing away from us. However, random selection is better than biased selection. Your strawman is introducing farside activity. Good to see you recognize random selection is inferior. Your description of bias selection is incorrect. Measuring every spot removes any bias along with any concerns of missing information.
|
|