Global warming

Miyazaki

ZING!

sidicks

And your point is?

And we should clearly do what we can to protect the natual environment - no argument there.

Seems sensible

I thought the 'science was settled' and we are set for certain doom if we don't stop buring fossil fuels?


I'm not sure anyone is suggesting that is the correct thing to do, at least I'm not.

Sidicks

sidicks

It's hardly being pedantic, it's a key point.

A greenhouse is pretty much a passive system so you can model the behaviour and accurately predict the response.

The earth and its atmosphere are very different, we don't currently have sufficient understanding or computer power to accurately model all the interactions and potential feedback loops, so we are making at best an educated guess as to what will happen in response to an increase in CO2 (particularly if the IPCC models choose to ignore some factors that might be expected to act as negative feedback loops.......).

That is why it is important to make it clear that the earth is NOT a greenhouse.

Again, more personal attacks, but I can live with that.  As it happens, I'm very successful in my profession.


Sidicks

sidicks

Explain???

Toko Black

Just for you this one time.

Your job models risk and makes recommendations based on predictions from those models.

People pay you and use that advice to make strategic decisions, set insurance levels.

Sound familiar ?

"The 'argument' will be won and lost based on the evidence of actual experience versus that predicted by the models"

sidicks

At a very simplistic level that describes the role that some actuaries perform, yes.

Obviously the results of those models then need to be considered in the commercial environment and a trade-off made between short-term profitability and long-term risk-return considerations.

Obviously the actual experience will nearly always differ from the modelled outcome, so contingency factor may be added.  Finally, there will also be some random variation that can't be modelled.

I'd say that generally actuaries involved in insurance pricing have been fairly successful on average, although the insurance cycle dicates that losses will be made at certain points in the cycle.

I don't work in insurance pricing though!

The key thing is that the models used have proved fairly accurate and if people don't like the prices then they don't have to buy the insurance....


Sidicks

Toko Black

And yet he still doesn't get it.

Someone pays you to make predictions from models so they cover themselves for the future.

Yet according to you:

"The 'argument' will be won and lost based on the evidence of actual experience versus that predicted by the models".

So doesn't that mean we should make you unemployed and instead wait to see what the outcomes are when they actually happen or not ?

EarthRod

I've learnt something new here.

Reading various articles on climate change has highlighted (so I thought) it is all based on statistics, historical measurements and current measurements. Then finally drawing and publishing conclusions on climate change based on the statistical information gathered.

A bit like forecasting the weather for the next 24 hours.

I did not know it's based on science.

sidicks

Exactly - an insurance company will soon go out of business if its models are wrong, either because it charges too little for the risk and therefore makes losses, or because it charges too much and gets no business.

If the projected claims experience is fundamentally different to that expected under the model, then the model is likely to be wrong and need amending.

Likewise if the actual experience for temperature change is significantly different than expected without obvious reason then the MMGW models should be reviewed and revised.

But they haven't!


That makes no sense!

Insurance modelling is about making projections for what might happen in the future, based on aLL known factors that can be modelled which affect claims frequency and amount.

MMGW modelling is currently about projecting future temperature changes based on CO2 emissions, and effectively assuming that CO2 is the only thing driving temperature. GIGO!

If climate modelling included all the key factors affecting temperature then the projections would be useful!

Insurance modelling has been proven to be fairly accurate, unlike climate modelling.


Sidicks

Toko Black

Then that is where you are going wrong

Forcasting the weather for the next 24hrs is nothing like prediciting the earths average temperature.

Imagine the difference between prediciting the average temperature increase or decrease in a bath as you add or remove hot water.
You can do basic calculations, add in heat loss through the surface of the water and the material of the bath.
It's not to complicated to work out and predict what the average temperature will be for the water in say 2 hrs.

However, if you try to predict the particular currents or eddies within the water and how temperature will change in one part of the bath to the next, you need to start using very complex models, fluid dynamics, and worrying about the movements of individual molecules of water.

Imagine trying to predict where an individual cloud will be if you know the approximate wind speed and direction .... then imagine trying to predict where a particular molecule of group of molecules will be within that cloud.

The local weather is those eddies and small groups of molecules in the bath or in that cloud, the global temperature is the average bath temperature or the position of the cloud.

Hope that helps shed some light on how massively different the two things are.