One way of filtering the original is to take x measurements together and take the average of x
x being 10, 30, 100 or something Problem with centuries is that we do not have any real data to fall back on. Most data for old temperatures is based on assumptions referred from something else. There is a tendency among many people to make assumptions and the try to find data that can help them prove their theories. The real problem is that you need some way of extracting the trend out of seasonal data. 2010/1/8 David Mitchell <[email protected]>: > One problem I have with the first graph from a purely technical point of view > is > that the number of horizontal data points far exceeds the resolution I used > for > the plot. So, the data as shown has been filtered by the underlying plot > routines. > > Here is a plot where the data actually fits the resolution: > > http://www.jsoftware.com/jwiki/Scripts/WeatherPlot?action=AttachFile&do=get&target=rochester2005temp.png > > I'm not if either plot tells me much about a phenomenon that most likely is > global and may be better analyzed on a scale of centuries. > > -- > David Mitchell > > On 1/7/2010 16:14, Björn Helgason wrote: >> Well your first graph looks interesting and it gives a picture of big >> variation of temperatures. >> So if you want to present any trend out of the data it needs to be >> treated with seasonal adjustments. >> >> Series are made up of four components: >> St: The Seasonal Component >> Tt: The Trend Component >> Ct: The Cyclical Component >> It: The Error, or irregular component. >> >> One way of looking at such data is take each daily value and compare >> that value with the average value for that day >> >> One scary thing about the discussions about heat measurements and >> trends over time is that the researchers have been manipulating the >> data to fit their interests. >> >> >> >> >> 2010/1/7 David Mitchell<[email protected]>: >>> Yes, the second data set is the minimum high and minimum low temperatures >>> for >>> yearly intervals. It is an extract from the original data. >>> >>> -- >>> David Mitchell >>> >>> On 1/7/2010 13:41, Björn Helgason wrote: >>>> The Y scale in one graph goes from -20 to +100 and so do the curves >>>> The Y scale on the other goes from -20 to +20 and the curves therefore >>>> not the same >>>> >>>> >>>> 2010/1/7 David Mitchell<[email protected]>: >>>>> It may depend on how one looks at the data: >>>>> >>>>> http://www.jsoftware.com/jwiki/Scripts/WeatherPlot?action=AttachFile&do=get&target=rochestermintemp.png >>>>> >>>>> ;) >>>>> >>>>> -- >>>>> David Mitchell >>>>> >>>>> On 1/7/2010 9:23, Björn Helgason wrote: >>>>>> Looking at this data it is nothing that says there is any general >>>>>> warming going on. >>>>>> >>>>>> If anything it looks like it is getting cooler. >>>>>> >>>>>> >>>>>> >>>>>> 2010/1/6 David Mitchell<[email protected]>: >>>>>>> >>>>>>> Here is one example of what I find useful with J: >>>>>>> >>>>>>> http://www.jsoftware.com/jwiki/Scripts/WeatherPlot >>>>>>> >>>>>>> -- >>>>>>> David Mitchell > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm -- Björn Helgason, Verkfræðingur Fornustekkum II 781 Hornafirði Po Box 127,801 Selfoss , t-póst: [email protected] gsm: +3546985532 sími: +3544781286 http://groups.google.com/group/J-Programming Tæknikunnátta höndlar hið flókna, sköpunargáfa er meistari einfaldleikans góður kennari getur stigið á tær án þess að glansinn fari af skónum /|_ .-----------------------------------. ,' .\ / | Með léttri lund verður | ,--' _,' | Dagurinn í dag | / / | Enn betri en gærdagurinn | ( -. | `-----------------------------------' | ) | (\_ _/) (`-. '--.) (='.'=) ♖♘♗♕♔♙ `. )----' (")_(") ☃☠ ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
