authority and credibility in dullish?

youll have to call paddypower to get the orges credentials.
4/1 odds in. waves fingers with chavs rings on all 5 fingers


On 08/12/2010 11:00 AM, Dawlish wrote:
On Dec 8, 10:46 am, Seany wrote:
On Dec 8, 10:15 am, wrote:



Will = Cold Winter Ramping
Dawlish = Global Warming Ramping

What's the difference?-

The only difference is that Will is a respected meteorologist who
speaks with authority and credibility. He also has a sense of humour.
Dawlish - ahh bless him!

I thought these comments might start as soon as I pointed out the
reality. *((

Thanks John

On 08/12/10 17:51, prodata wrote:
On Dec 8, 3:19 pm, wrote:

I've calculated the December mean from 1971 - 2009 as +5.0c, with a
Standard Deviation of 1.5c, so by my calculations the current CET that
is running at -2.0c is somewhere between 4& 5 Standard Deviations.

Is there evidence that these values are normally distributed? Maybe
this has been throughly tested and is well-known? But if not then
personally I'd be happier with a non-parametric analysis.

JGD

Technically speaking they won't be, as the normal distribution is
unbounded, whereas there is a limit on how large a magnitude a
temperature anomaly can be. However for practical purposes it is likely
that temperature anomalies will be close enough to normal that the usual
statistical techniques that require normality will be valid.

On Dec 8, 11:11*pm, Adam Lea wrote:
On 08/12/10 17:51, prodata wrote:

On Dec 8, 3:19 pm, *wrote:

I've calculated the December mean from 1971 - 2009 as +5.0c, with a
Standard Deviation of 1.5c, so by my calculations the current CET that
is running at -2.0c is somewhere between 4& *5 Standard Deviations.

Is there evidence that these values are normally distributed? Maybe
this has been throughly tested and is well-known? But if not then
personally I'd be happier with a non-parametric analysis.

JGD

Technically speaking they won't be, as the normal distribution is
unbounded, whereas there is a limit on how large a magnitude a
temperature anomaly can be. However for practical purposes it is likely
that temperature anomalies will be close enough to normal that the usual
statistical techniques that require normality will be valid.

IMHO, the normal distribution is for random values but weather/climate
is chaotic. Chaos can appear random but is actually deterministic.
The fact that the temperatures do not fit happily into a normal
distribution only goes to prove that they are not random.

The classical case of a chaotic attractor is the Butterfly effect
produced by the meteorologist Edward Lorenz.
http://www.viewsfromscience.com/docu.../chaos_p3.html
It does not represent a plot of daily temperatures over the years, but
if it did, then you can see that much of the time there it is close to
a cycle just like annual temperatures. However, at times it jumps out
of that cycle into another cycle. What we could be seeing with the
December temperatures being so far from the standard deviation that
the climate is jumping out of the current cycle/state/attractor into
another climate state, i.e. an abrupt climate change. Let's hope not.

OTOH, what we may be seeing is are the effects of an abnormal sum. The
last solar minimum lasted much longer than usual.

Will mentioned in another thread that it may be due to there being
less Arctic sea ice, but the area is not much different from other
recent years. It could be that the ice is thinner allowing more leads
to form and so allowing more water vapour to escape from the sea
beneath the ice. In that case, could that be the reason for low
pressure over the pole, and equalising high pressure over the North
Atlantic?

Cheers, Alastair.

Cheers, Alastair.

On 2010-12-08 20:10:13 +0000, Teignmouth said:

John,

If I use the SKEW function I get the following values:

Period Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1659-2009 (0.6) (0.5) (0.1) (0.1) 0.1 0.2 0.4 0.3 0.0 (0.1) 0.0
(0.2)
1971-2009 (0.8) (0.8) (0.2) 0.2 (0.1) (0.2) 0.8 0.4 (0.1) 0.0
(0.1) (0.8)

Now what do I do with the value and the Standard Deviation?

If I have a December mean for the period 1659-2009 of +4.1c, and the
STDEV is 1.7c, and the SKEW is -0.2c, do I get a revised STDEV of 1.5c
or 1.9c? Then do I use +4.1c and the revised STDEV x1 x2 x3 etc to
get the revised Standard Deviation thresholds?

Your help is much appreciated.

Thanks

No, just use the standard deviation. The skew is tiny. It's as good a
normal distribution as you'll ever get.

There are tests for the significance of deviation from normality, but I
wouldn't bother with them here. What do you want to do with your
standard deviation? Convert the current mean to a z-score?

--

Trevor
Lundie, near Dundee
www.trevorharley.com

In article 2010120908504475249-taharley@dundeeacuk,
Trevor Harley writes:
On 2010-12-08 20:10:13 +0000, Teignmouth said:

John,
If I use the SKEW function I get the following values:
Period Jan Feb Mar Apr May Jun Jul Aug
Sep Oct Nov Dec
1659-2009 (0.6) (0.5) (0.1) (0.1) 0.1 0.2 0.4 0.3 0.0 (0.1) 0.0
(0.2)
1971-2009 (0.8) (0.8) (0.2) 0.2 (0.1) (0.2) 0.8 0.4 (0.1) 0.0
(0.1) (0.8)
Now what do I do with the value and the Standard Deviation?
If I have a December mean for the period 1659-2009 of +4.1c, and the
STDEV is 1.7c, and the SKEW is -0.2c, do I get a revised STDEV of 1.5c
or 1.9c? Then do I use +4.1c and the revised STDEV x1 x2 x3 etc to
get the revised Standard Deviation thresholds?
Your help is much appreciated.
Thanks

No, just use the standard deviation. The skew is tiny. It's as good a
normal distribution as you'll ever get.

Surely a skew of -0.8, as for Decembers over the last 40 years, is
enough that it ought to be taken into account? It's interesting that the
winter month skew values have been larger over the last 40 years have
been larger that when one considers the whole CET record.

snip
--
John Hall
"I look upon it, that he who does not mind his belly,
will hardly mind anything else."
Dr Samuel Johnson (1709-84)

On 08/12/10 20:10, Teignmouth wrote:
John,

If I use the SKEW function I get the following values:

Period Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1659-2009 (0.6) (0.5) (0.1) (0.1) 0.1 0.2 0.4 0.3 0.0 (0.1) 0.0
(0.2)
1971-2009 (0.8) (0.8) (0.2) 0.2 (0.1) (0.2) 0.8 0.4 (0.1) 0.0
(0.1) (0.8)

Now what do I do with the value and the Standard Deviation?

If I have a December mean for the period 1659-2009 of +4.1c, and the
STDEV is 1.7c, and the SKEW is -0.2c, do I get a revised STDEV of 1.5c
or 1.9c? Then do I use +4.1c and the revised STDEV x1 x2 x3 etc to
get the revised Standard Deviation thresholds?

Your help is much appreciated.

Thanks


It is possible to calculate a standard deviation for a skewed
distribution using the distribution parameters. This requires you to
know which distribution best fits the data.

e.g. Gamma distribution: http://en.wikipedia.or/wiki/Gamma_distribution
variance=k*theta^2 where k and theta are parameters that define the
distribution.

On 2010-12-09 10:10:26 +0000, John Hall said:

In article 2010120908504475249-taharley@dundeeacuk,
Trevor Harley writes:
On 2010-12-08 20:10:13 +0000, Teignmouth said:

John,
If I use the SKEW function I get the following values:
Period Jan Feb Mar Apr May Jun Jul Aug
Sep Oct Nov Dec
1659-2009 (0.6) (0.5) (0.1) (0.1) 0.1 0.2 0.4
0.3 0.0 (0.1) 0.0
(0.2)
1971-2009 (0.8) (0.8) (0.2) 0.2 (0.1) (0.2) 0.8
0.4 (0.1) 0.0
(0.1) (0.8)
Now what do I do with the value and the Standard Deviation?
If I have a December mean for the period 1659-2009 of +4.1c, and the
STDEV is 1.7c, and the SKEW is -0.2c, do I get a revised STDEV of 1.5c
or 1.9c? Then do I use +4.1c and the revised STDEV x1 x2 x3 etc to
get the revised Standard Deviation thresholds?
Your help is much appreciated.
Thanks

No, just use the standard deviation. The skew is tiny. It's as good a
normal distribution as you'll ever get.

Surely a skew of -0.8, as for Decembers over the last 40 years, is
enough that it ought to be taken into account? It's interesting that the
winter month skew values have been larger over the last 40 years have
been larger that when one considers the whole CET record.


I did the descriptive stats for December since 1659 in SPSS.

1659-2009
Mean 4.08
Median 4.10
sd 1.72
Skewness -0.25
SE of skewness 0.13
Skewness Z = -1.92
Kurtosis -0.92
SE kurtosis .26

So there's no kurtosis (bumpiness) and the z-score for skewness isn't
(quite) significant either. (A negative skew means too long tail on the
left side.) But my understanding is that with large samples (like this)
you shouldn't worry too much about significant skew becaus the standard
error of the skew tends to be small. So when you've got a large sample
(over 200) you need a z score giving a p of at least 0.01 before you
need to start to worry.

So as you would expect the complete series is normally distributed.

From 1971-2009,

Mean = 4.97
Median = 5.3
SD = 1.5

Skew = -0.843
SE skew = 0.378
Z skewness = 2.23

kurtsosis = 1.52
se kurtosis = 0.741
Z kurtosis = 2.05

Both of which are significant 0.05p0.1.

Having said that, I wouldn't worry about it too much. If you're really
concerned you could transform the data, but the ones I'm familiar with
only work for positive skew. I've been told you can add a constant to
the mean and then apply a logarithmic transformation, but I'd be
surprised if it really makes a difference.

I'm no statistics expert; just a psychologist who uses them.

--

Trevor
Pedagogical in Lundie, near Dundee
http://www.personal.dundee.ac.uk/~taharley/

In article 2010121010093916807-taharley@dundeeacuk,
Trevor Harley writes:
snip
I'm no statistics expert; just a psychologist who uses them.

Nevertheless that was pretty impressive. Thanks for going to all that
trouble, Trevor. I'm sure that Teignmouth will find your results useful.
--
John Hall
"I look upon it, that he who does not mind his belly,
will hardly mind anything else."
Dr Samuel Johnson (1709-84)

On Dec 9, 12:40*am, Alastair wrote:
On Dec 8, 11:11*pm, Adam Lea wrote:





On 08/12/10 17:51, prodata wrote:

On Dec 8, 3:19 pm, *wrote:

I've calculated the December mean from 1971 - 2009 as +5.0c, with a
Standard Deviation of 1.5c, so by my calculations the current CET that
is running at -2.0c is somewhere between 4& *5 Standard Deviations..

Is there evidence that these values are normally distributed? Maybe
this has been throughly tested and is well-known? But if not then
personally I'd be happier with a non-parametric analysis.

JGD

Technically speaking they won't be, as the normal distribution is
unbounded, whereas there is a limit on how large a magnitude a
temperature anomaly can be. However for practical purposes it is likely
that temperature anomalies will be close enough to normal that the usual
statistical techniques that require normality will be valid.

IMHO, the normal distribution is for random values but weather/climate
is chaotic. Chaos can appear random but is actually deterministic.
The fact that the temperatures do not fit happily into a normal
distribution only goes to prove that they are not random.

The classical case of a chaotic attractor is the Butterfly effect
produced by the meteorologist Edward Lorenz.http://www.viewsfromscience.com/docu.../chaos_p3.html
It does not represent a plot of daily temperatures over the years, but
if it did, then you can see that much of the time there it is close to
a cycle just like annual temperatures. However, at times it jumps out
of that cycle into another cycle. What we could be seeing with the
December temperatures being so far from the standard deviation that
the climate is jumping out of the current cycle/state/attractor into
another climate state, i.e. an abrupt climate change. Let's hope not.

OTOH, what we may be seeing is are the effects of an abnormal sum. The
last solar minimum lasted much longer than usual.

Will mentioned in another thread that it may be due to there being
less Arctic sea ice, but the area is not much different from other
recent years. It could be that the ice is thinner allowing more leads
to form and so allowing more water vapour to escape from the sea
beneath the ice. In that case, could that be the reason for low
pressure over the *pole, and equalising high pressure over the North
Atlantic?

Cheers, Alastair.

Cheers, Alastair.- Hide quoted text -

- Show quoted text -

As a follow upt to what I wrote above, can I menton that there as a TV
program on BBC 4 last night entitled "The Secret Life of Chaos" which
can be seen for the next 6 days at:
http://www.bbc.co.uk/programmes/b00pv1c3

It says that weather is Chaotic, and therefore it is detrministic and
not random. In that case the normal distributions does not apply to
weather data.

Cheers, Alastair.