Previous  1 ... 16  17  18  19  20  21  22 ... 24  Next

You're making a fool of yourself - someone actually requested to see what you have to say, and you're writing nonsense.

Explain your so-called proof:


Your claim is that this single graphic that I posted (in a discussion unrelated to the data itself):

Misrepresents data to demonstrate a trend, but an opposite trend is obvious.

- what misrepresentation of the data used to form this graph (whatever that may be, since I didn't provide it as a source) occured?

- what trend, opposite to the one shown, is obvious from this graph?

You made the claims - back them up. Anything but a serious response to these questions is an admission of your cowardice.

   

Bluenose

The only thing this graph with any accuracy depicts is the turn of the century trend. The rest of the graphs profile is entirely fictitious. Where are the warming trends of the 20-30s which approximated late 90's values. It is just Goron science fiction.
Bluenose


Not really I am attempting to direct you to enlighten yourself and once and for all have an epiphany on the way to KYOTO. I have no doubt that others have grasped this loose end and pulled but you just obtusely demand to be spoonfed---that's cowardess.

Specifically the reasons why even the IPCC abandoned it's preposterous claim that the Atlantic thermohaline circulation could be stopped.

Now there are a few more strange words you will have to deal with.

   

Answer the questions above and enlighten me, then.

You made the statements with respect to that specific graph - stand up for them.

Unless you're a coward, then you're off the hook.

   

Read my post and then go deepen your shallow knowledge on the matters I specified.

I can't enlighten you----that's your job.

Unless you are a total, unrepentant ignoramus.

   

You keep adding things to your posts after I replied to them - stop that. I'm not concerned with any other matter than the one, simple statement you made in reference to that one, simple graph.

I've explained my reasons for challenging this explicitly. Unless you can defend your argument with something substantial - you've yet to do anything of the sort - then you're backing away from any real discussion.

You're a coward.

   

Say Jökulhlaups.

   

Well I've reduced you to inane rambling. Have a nice day.

   

Don't pat yourself on the back too hard, there, his posts are rarely more than that to begin with...

   

I recall the 60's the little communist idiots would leave the Field (so to speak) shouting their stupid slogans.

Say Jökulhlaups.

   

and "Say Jökulhlaups" is your stupid slogan?

Your analogy fits very well in this case. Weird that you'd call yourself a little communist idiot, though.

   

Blue_Nose Posted:

Better than you could possibly imagine.


Say Jökulhlaups

   

BN I have some questions as regards anomaly graphs. Honest to God. I'm not trying to make a point at this time, or even leading up to one. I've done, that and I'm quite satisfied with the point made. What I'm requesting here is just knowledge. You know more about qraphing than I do.

OK, so here's what I want to know. With anomaly graphs such as the one I got from Zipperfish's link seen below, how exactly do they do the averaging of anomalies to get the curvy line? Do they average out high and low anomalies for a specific time period? That's the impression I got from the bits of reading I've done.




For example with this one I see they've created lines for monthly, annual, and 5 year periods. Why is it on some anomaly graphs they don't seem to offer the time period for the averaging? Wouldn't that be important? Is there a generally accepted time period over which averages are done from point to point?

   

While I would have prefered that you address the points I made in response to your last post, any actual discussion of actual topics beats what Bart and samsquanch have offered. I'm going to trust that you're sincere with these questions.

There are a number of ways to treat trend fitting - in this case, where's there's no outstanding, consistent trend, the averaging technique - known as Simple Moving Average - you described is the most suitable. It's the same trend-finding method they use in the Stock Market.

That would consist basically of averaging the values within a certain timespan to obtain an average at each time.

There are weighted averages as well, which consist of assigning a higher weight to more recent data, to allow the curve to adapt more quickly to rapidly changing data.

There's room for adjustment as the person doing the analysis sees fit - if the trendline is too shakey, they may use a longer timespan to smooth it out even more, etc.

That second graph, it's laid out for you there exactly as you see it - a 5 year trend consists of the average over 5 years, annual over a year, etc.

In that case as well, it appears as though the average is taken over a range centered about the point in question - if you look at the last available 5-year trend point, it occurs three years before the end of the data. That is a clear indication that the five years ranged from two years previous to two years ahead.

Again, if they use a weighted system, that's not indicated, but they probably didn't.

edit: oops, you asked why that's not as obvious in other graphs, not that one.

Well, the trend is simply a tool used to show a trend in noisey data. I don't see how knowing the precise details of that trend-fitting is important knowledge, since it's simply a visual aid, not actual data.

   

Well, I wouldn't count on that. He gets pretty snippy once it all starts to fall apart for him.

   

Thanks BN. That's what I wanted to know.

On your points concerning my post, I didn't think those points required response. As I saw it, I made a point, and you critiqued it. That seemed fair. I saw the questions as largely rhetorical, but if you honestly have something you want to ask. Feel free. I'll do my best to answer your questions just as you did.

   

Previous  1 ... 16  17  18  19  20  21  22 ... 24  Next