DIY data?

[dennis]

You still haven't got the idea yet, I can reel off a whole list of numbers and it would still be wrong just as your average is. Or you could be boring and halve 19 and add it to -14.

[Roger Chapman]

I see you are back in full wiggle mode.

You claimed the error above was because you miss keyed a 9. But which figure was that? The significant figures you quoted were maximum 5C, minimum -14C and average -8. So which one of those was supposed to be a

9 in *your* faulty data set?

Oh, and while you are it why does your pitiful lack of ability in the Maths department mean I "still haven't got the idea yet". You missed the chance to pretend that you introduced a deliberate error to catch me out when you opted for the miss key excuse instead.

[dennis]

It wasn't a deliberate error just a typo, thats what you get when you have drunk a bottle of wine. You should see TNP posts when he has been on the booze, there are only a few words with the correct number of letters let alone to correct ones. However it serves to show the point, your average is wrong, just as wrong as my typo. The difference is I know its an error while you don't have a clue.

[Man at B&Q]

It's not just trees

but aircon exhausts and all kinds of urbanisation that has affected many stations.

MBQ

[Man at B&Q]

So what do you think the average is? Why?

I can come up with a number of very different averages, all equally valid depending in the sample timing and frequency.

Which one is "correct"?

MBQ

[dennis]

How hard can this be?

He has two points on an curve with an unknown shape. How can he know the average of that curve?

With a random(ish) curve like weather you can be more precise the more samples you have but having just the max and the min to average isn't going to get you a point on the graph as accurately as having a value at a defined time like 9 am.

Obviously the one that "proves" GW.

All I can say is I hope the climatologists don't make assumptions the way Roger does or we will have to throw away everything they have ever done and start again.

[Roger Chapman]

No wonder you drive the way you do.

But come on, don't be coy. Be a man and say which of the three numbers you wish to replace with a 9 to arrive at your intended message. I am inserting below the paragraph you silently edited out as a reminder of the simplicity of the action required of you.

"You claimed the error above was because you miss keyed a 9. But which figure was that? The significant figures you quoted were maximum 5C, minimum -14C and average -8. So which one of those was supposed to be a

9 in *your* faulty data set?"

The TNP is well known for his incomprehensible typing. Whether it is because he is drunk most of the time is just your assumption and, as it is you, probably a wrong assumption.

My average is an approximation. Approximations by definition are not precise.

So you say but I have long since learned never to take anything you say on trust.

[Roger Chapman]

He says one of those figures should be a 9 but is being coy about which one.

Not just any two points though, just the two most significant.

Never mind the temperature, feel the time.

We? I am absolutely confident that none of us are going to be included among the experts who are now being forced to go over the whole database in great detail.

[dennis]

Well yes I do only drive when sober, not even a shandy if I am driving.

You can add two numbers together and divide by two as easily as anyone else can, this is a DIY group.

It is your assumption, I never said anything about how often or how long he may be drunk for and as you say you are probably wrong about that too.

[dennis]

Why aren't you? You are expert enough to make assumptions and claim they are good assumptions.

[Roger Chapman]

I am expert enough to see through a fraud like you.

[Man at B&Q]

[e-mail]...

As hard as you want to make it.

I was referring to *your* hypothecated data.

He can know the average of two samples.

If we change the defined time to midnight, is that better or worse?

Which is the more *useful* figure?

I wasn't looking for a facetious answer. But, if that's all you're capable of...

MBQ

[Roger Chapman]

Isn't that what that disgraced German bishop used to say?

Oh yes I can add two numbers together, divide by two and even get the right answer, unlike you. But which two? Where do I have to substitute that missing nine in order to confirm how you fouled up?

Let us see:

The average of 5 and -14 is -4.5 so the 9 doesn't fit there.

If the upper limit is 5 and the average -8 then the lower limit is -23 so that isn't the typo either.

If the lower limit is -14 and the average is -8 then the upper limit is

-2 so not that either.

Conclusion - not just a simple typo then.

This is getting tedious so how about the average of -5 and -14.

Golly gosh, what a surprise. There at last is a 9. (OK actually 9.5 but one can hardly expect such accuracy from someone who can't cope with negative numbers).

No wonder Dennis is trying to wriggle his way out of admitting to such a stupid error.

snip

[dennis]

messagenews: [e-mail]...

Which is *not* the average temp for the day.

It makes no difference.

As I stated before you can work out the yearly (or any other period like months, 27.8 days, etc) average of the Max temp, min temp and 9 am temp, you can't workout the yearly average temp of the day.

As I already stated, the 9 am time is absolute, you can work out the yearly average 9 am temp, this is accurate unlike averaging the min and max to get the "average" day temp.

It is obviously better if you have accurate figures when you are looking for minor changes, making assumptions that make bigger difference than what you are looking for is bad.

[dennis]

You still can't grasp that its no more an error than your assumption and that I could have put any figure in there as long as it was between max and min and it was as likely to be the real average as your guess. I guess you aren't good enough for sarcasm.

[Man at B&Q]

[e-mail]...

It's *an* average of the daily temp. Whether its *the* average is another matter.

That doesn't answer the question.

MBQ

[Roger Chapman]

First what Dennis didn't want to repeat with markers added to place in sequence:

You can wriggle all you like Dennis but you have been well and truly nailed as a liar and a fraud. There was no way that mistake was ever going to be the typo you claimed it was and just editing out the evidence to try and reduce the exposure won't work.

[dennis]

No its an average of the max and min. It has nothing to do with being an average of the daily temp.

#>> > Which is the more *useful* figure?

It damn well does. The only reason for choosing an erroneous figure is to falsify the data and you called that factious.

I am not going to repeat the arguments so it may as well end here.

[Man at B&Q]

[e-mail]...

It's an average of two reading of daily temperature. More readings would obviously be better. On most days, without violent swings in temperature, the result is probably close enough to use for spotting trends.

No. The question was "which is the more useful?". You didn't answer that. As usual you just wriggled.

Are you getting fractious? The word was facetious.

Realised you have no argument, you mean.

MBQ

[dennis]

messagenews: [e-mail]...

I have answered the question! The fact that you don't understand the answer is your problem, its pretty obvious which is more useful from what I have been saying.

Tell that to the spell chucker not me.

Obviously no argument that is down to a level where you understand it.