Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.



Re: a problem with temperature records
Posted:
Apr 13, 2013 5:35 PM


On 13.04.2013 15:00, josephus wrote:
> I have some data sets ENGLAND 17291970, NOAA 18802012 GISS == > NOAA data. I even have a data set for ICE but that is not my > problem. > > the problem is when I run a least squares algorithm over the data > set. should I change the extent, the slope increases (??). > > so I plotted the aggregate list of slopes // mod X. then I plot that > list. and this is were the problem exceeds my programming and > mathematics. I can take slope data at any scale (all slopes are > 2012 to N). I take all my slopes upward If I reversed the process I > would expect to get negative slopes. > > > this is a plot of the slopes over the data NOAA and excluding data > set downward slopes. this is an artifact of GW. the ranges of the > data are increasing together. the result is my BEST FIT line also > is > positive and increasing. > > the big question is WHAT KIND of PROCESSING caused the RAW DATA to > show irregular slopes like these > > If you want I can send you the RAW data and the slope list. > > this is a much more complex problem than just displaying odd data. > every data set I have processed agrees with the general terms of this > program they all plot upward except for ENGLAND it slopes downward > > I believe this data set was conformed by a process. but the > artifacts > have strange properties it may be a trap door function, but I am > interesstd in understanding the slope data that I have produced. > > josephus > > > MY regular plot [scatter gram] ...
I tried to understand you question  but I am a bit lost (may be it is my English, and I do not understand you abbreviations).
Could you provide your data (by a link?) and clearly say what you do or expect or want? It is not clear what "least squares algorithm over the data set" should be, or "negative slopes".
Do you want some approximating 'curve' to your data? Or just a 'line'? Or what?



