How do you measure uncertainty?

[Lying Toad]

The toad was recently asked by a scientist:

?How do you measure uncertainty?? I responded with ?fixed and variables?. This is an everyday question we ask ourselves to quantify the day ahead, our security, through to our future plans.

She then expanded and said that this is a question she askes Ministers when they consided new plans.....

How do you measure uncertainty?

TLT

[HAL9000]

Start here and follow the links: Uncertainty - good luck!

[SeanMacGabhann]

I'm not sure

[HAL9000]

^That made me laugh ? brilliant!

[keano77]

Hmmm, interesting one this.

Multidisciplinary.

I suppose the answer to your question is you can't measure uncertainty. To measure implies a known, or agreed, standard of measurement. Uncertainty, by definition, can't be quantified. Prima facie, the question is illogical.

BUT, science tells us a cat can be both alive or dead at the same time depending on the observer and that an electron can be in two places at the same time.

So the question isn't as daft as it first appears.

However I'd advise your scientist to keep ministers out of the debate otherwise therein the road to madness lies

[Ladymuck]

> I suppose the answer to your question is you can't

> measure uncertainty. To measure implies a known,

> or agreed, standard of measurement. Uncertainty,

> by definition, can't be quantified.

You mean like infinity?

[Chick]

When did "science tells us a cat can be both alive or dead at the same time depending on the observer"?

[KidKruger]

Chick, look up Schroedinger's Cat.

[KidKruger]

I think that Infinity, by definition, IS a measurement, though not one that has a set numemical value but more a property or behaviour that is defined. ie. never ending and therefore perpetual, a bit like mathematical limits where you don't get an exact number as the answer but a 'tendency' towards a sepcific (range of) number(s).

[HAL9000]

As soon as Schr?dinger's cat observes Chick - that's lunch sorted. No uncertainty there! (I taut I thaw a puddy tat!)

Seriously, if you do look up Schr?dinger's cat you have a 50:50 chance of finding it dead - best not go there.

Sadly - attempting to measure uncertainty has killed more cats than any other scientific experiment. Be nice - don?t open the box!

[Lying Toad]

Using a 'Starting Point' is an automatic action by us all - and starts at 50-50.

1. When you cycle through a GREEN traffic light at a 4-way junction you are pretty certain you will not be hit by a car (on your Left or Right hand side) as they are at a RED light. But what stops them, as a RED light is purely a symbolic function?

2. However, the same set of lights, a pedestrian crossing when the green man appears (and I don't mean from Mars) stands a greater chance of being hit by a cyclist, yet..... its the same symbolic function?

Unfortunately, more incidents with cyclist will help with this measurement of uncertainty.

TLT

[John247]

How do you measure uncertainty? Uncertainty is what's left after you've measured everything else...

[louisiana]

First, you get some training in statistics, so that you can understand and apply probability theory, statistical mechanics, decision theory etc.

The scientist's question was rhetorical: she will of course have a thorough grounding in statistics, probability, risk, uncertainty... as part of her scientific training. The problem with most politicians is they don't have a clue. So the question is an interesting one to ask, to determine whether they have any sound framework at all for arriving at decisions.

[Ladymuck]

> Unfortunately, more incidents with cyclist will

> help with this measurement of uncertainty.

Wouldn't different categories of uncertainty be subject to different methods of measurement?

To me, there are different classes of uncertainty. E.g., I view the uncertainty alluded to immediately above as being of the variety which becomes more certain with experimentation. Whereas, where in the OP you talk of "fixed and variables", this type of uncertainty would become more certain with the application of probability and chance? No?

[????]

Can you be certain about certainty. Does F=MA?

[Ladymuck]

Haven't a clue. And of that you may be 100% certain!

[????]

don't start winding me up....(6)

[Ladymuck]

100 + 100 + 500 = a wound-up Master Quids LOL

[????]

I'm certain we're in for a telling off from the Chair anytime sson

[Ladymuck]

You may be right LOL...

Sorry esteemed Chair: back on topic...

[louisiana]

I'm just trying to get the tape measure around the tongue, in the cheek, as it were...

[louisiana]

I'm sure the Chair would be amenable to a spot of Middle Eastern-style corruption. After all, that is the way the world works, as Drawing Room afficionados will be aware... [gets out her decision tree and adds some probabilities to various branches]

[louisiana]

Heisenberg.

[HAL9000]

\\|||//

/ ~ ~ \

) O O (

-------uUu------U------uUu-------

Heisenberg may have been here

 

[fyvum]

\\|||//

/ ~ ~ \

) O O (

-------uUu------U------uUu-------

Heisenberg may have been here

That is a pretty good likeness of 'Chad' if my memory serves Hal, but Heisenberg sounds like a lager.......