(Before I go on a bit of required reading:)
In no particular order
- Possum's investigation This has links to the Hawke report, Audit commission et al as well.
- Possum on CSIRO report )
One thing I didn't understand was how pink batts could cause problems with potential fires. In fact they didn't!
The problem was where contractors actually poured installation stuff ( highly technical term) into the roof and it got ,as one might expect, very very close to downlights ( which get quite hot if left on for a reasonable time.) When the downlights were left on for a considerable period then fires eventuated.
I thought I might just make a point no-one else has now.
If you look at the number of fires before the program. there were ABOUT 85 fires from ABOUT 65,000 installations of insulation. ( I love that!)
Now if that ratio carries through to 1.5 million installations then you get over 1960 fires. What actually happened is that we got almost ten times less.
You do not have to do a z-test to see there is a large difference here although it will tell you they are significantly different!
Now I am not saying there were no problems with this program. There were as the Audit office points out but two things stand out like the proverbial shag on a rock.
- A lot of houses are using less energy then they were before the program
- Safety in the industry improved enormously. Indeed if installation was actually physically installed alah pink batts and were not allowed to be poured in the roof then the number of fires would have been minimal.
Idiot
ReplyDeleteThe fires were the direct result of the program. In other words, no program no fires.
Your stated conclusion is that the program caused less fires, which is so stupid it defies anything you've previously said was dense.
write like that again and the comment will be deleted.
ReplyDeletewow if there is no insulation there are no fires!
The major point is there were about 10 times LESS fires under the scheme. Moreover if the insulation was limited to pink batts or something similar which must be physically put it there would have been very very few, if any.
Only a statistical ignoramus would argue otherwise
"Only a statistical ignoramus would argue otherwise"
ReplyDeleteYou mean like you suggested, that the program caused less fires?
200 homes burnt down as a result of the Lurch/Rudd insulation fiasco.
What part of the english language don't you understand?
ReplyDeleteThe number of fires were about TEN times less.
It is statistically significant!
If you wish to debate then use some evidence.
This is catallaxy!
"The number of fires were about TEN times less".
ReplyDeleteNo they weren't. You're lying.
Mark,
ReplyDeleteeither show why the figure is wrong or go away.
Last warning.
If the industry standard stayed the same there should have been over 1960 fires. There weren't.
do a z test o both the figures!
make that on both the figures
ReplyDelete"Mark,
ReplyDeleteeither show why the figure is wrong or go away.
Last warning."
Last warning? Huh. I'm the only commenter here.
I don't know how else to tell you. The fires were directly attributable to the Lurch/Rudd insulation program. No Lurch/Rudd program and those homes would not have burnt down. That's final. You're deeply deeply confused.
1)The wording is what Mark uses.
ReplyDelete2)No, no insulation no fires. All the program did was vastly increase the number of houses that were insulated however the number of houses that were affected by fires was vastly less.
"All the program did was vastly increase the number of houses that were insulated however the number of houses that were affected by fires was vastly less."
ReplyDeleteStop the self delusion.
Assertions without evidence is delusional!
ReplyDeleteThe argument of the anti group, that insulation kills, is ludicrous.
ReplyDeleterog
Great, intellectual giant Rog is onto this.
ReplyDeleteThis comment has been removed by a blog administrator.
ReplyDeleteThis comment has been removed by a blog administrator.
ReplyDeleteDo not swear here.
ReplyDeleteThere is a relationship between the marginal and average effect.
If the average is falling sharply then we know the marginal is also falling as it is here.
A person who doesn't understand either could believe the opposite but they would be very stupid.
If a Z test shows there is a statistical significant difference between the two series then a sensible person should start thinking clearly.
"If the average is falling sharply then we know the marginal is also falling as it is here."
ReplyDeleteThis is an abuse of economics.
You are simply presuming that the average is "falling sharply" then go back to your predetermined conclusion.
I hope you do not teach economics anymore.
You simply do not understand marginal and average effects and so fall for "Possum's" bizzare and extremely wayward "analysis".
"If a Z test shows there is a statistical significant difference between the two series then a sensible person should start thinking clearly."
No Homer.
Why don't you understand the following graph is simply damning:
http://blogs.crikey.com.au/pollytics/files/2011/04/firesper10dayperiod.png
Why don't we see a similar graph for new homes catching on fire, or air conditioner installations causing a spike in fires 0-10 and 20-20 days after installation?
"The major point is there were about 10 times LESS fires under the scheme. Moreover if the insulation was limited to pink batts or something similar which must be physically put it there would have been very very few, if any."
You simply do not understand average/marginal effects like you think you described above. You are saying for all that needed to happen to vastly improve safety (fires, not necessarily deaths), four people had to die.
If you average one death per house fire, only about 1/40 house fires were lethal before that time.
Please retire.
We know the average numbers. We know the number of fires fell drastically after the scheme was implemented.
ReplyDeleteThat is the average number fell like a rock.
Hence we do know what would have happened to the marginal number.
It could NOT have risen. It could NOT have even fallen slower.
It fell faster, a hell of a lot faster
Mark, do not use jargon you have no idea about
"That is the average number fell like a rock.
ReplyDeleteHence we do know what would have happened to the marginal number.
It could NOT have risen. It could NOT have even fallen slower."
No. You are just reparameterising your assumptions until you get the right conclusion.
Please retire Homer.
I'll recant my rather harsh judgment of you if you explain why it is tolerable to have a 20 day post installation period where pink batts burn down homes...but that doesn't happen with ordinary home installations.
As for using jargon you don't understand, you parroting Quiggin and telling everyone to ignore autocorrelation and cointegration in time series data was a laff riot.
Read Mark,
ReplyDeletePink batts rarely burnt any house down.
We know what happened to the average numbers because we have them. Given the relationship between average and marginal we can infer that marginal would be falling.
Feel free to demonstrate the opposite.
Mark you did not even know how many observations you need to know whether something is significantly different or not.
You parrot jargon without knowing what you are saying.
I quoted both people. I said very little actually.
Thank you again for showing your catallaxy mind
The problem with the HIP was not the insulation per se it was lack of regulation. The anti group are not keen on regulation so avoid that particular item. On its own insulation does not kill.
ReplyDeleteRog
Yes rog and it was the type of insulation.
ReplyDeletePink Batts as I wrote were rarely a problem it was the foamy stuff they poured in which was the problem.
You intellectual cumquat Rog. It was a huge government program in the first place that pushed demand for insulation to unseen levels and subsequently attracted marginal and shady operators to move in. If the government suddenly created artificial demand you have to watch out for consequences.
ReplyDeleteYour stupidity in matching the Lurch Rudd insulation fiasco with opinions on regulation is about as pathetic as they come.
You are far too dense to even post on Homer's blog and that's saying something.
Of course Homer agrees with you
Anymore language like that and the comments wil be deleted.
ReplyDeleteYes of course it pushed demand to much higher levels.
That was the point.
However fires reduced dramatically. That is the other point!
If you could show otherwise you would have.
You cannot hence the language
"I'll recant my rather harsh judgment of you if you explain why it is tolerable to have a 20 day post installation period where pink batts burn down homes...but that doesn't happen with ordinary home installations."
ReplyDeleteAnswer the question. Be a man.
Read, please read,
ReplyDeletePink batts rarely are involved in houses catching fire.
Foam my friend, foam that's the reason
Not good enough Homer.
ReplyDeleteWhy did these insulation related fires happen at all, when most houses never see a fire in their lifetime - let alone from post installation of any foam or insulation related material - as a new build or as renovative or restorative work?
If you had any truthfulness to your claims, the real test of statistical significance should be new builds vs pink batts and pink batts vs insulative work at any time outside of the stimulus.
You won't run these tests as it is above your ken and you know the results are unfavourable.
Mark,
ReplyDeleteIf you had read anything you would know at least one of those tests are not possible.
Why did they happen?
I have already explained that.
Don't run tests.
You make claims about the marginal effects and have NEVER done any work to support your 'thesis'!
"If you had read anything you would know at least one of those tests are not possible."
ReplyDeleteProve it Homer. What theory. What statistical theory says this is impossible?
Maybe you just don't know what you're doing.
Then Mark you show where anyone has provided evidence of how the different ways of insulating a house has been documented to do a test
ReplyDelete"Then Mark you show where anyone has provided evidence of how the different ways of insulating a house has been documented to do a test"
ReplyDeleteGibberish. You have tried to prove something with an assertion and no falsification. This is contrary to the spirit of statistical inference and testing.
You could prove the pink batts did or did not make houses more or less safe with intervention testing using VAR analysis Homer.
See Enders, W. Applied Econometric Analysis, 2010, Wiley, 3rd edn., for more detail.
Please Homer, tell me again why it is "impossible"? it isn't, is it, you just made that up because you don't know what you are talking about, do you?
Mark, you have aspergers as no-one can be that dense after being told why you cannot measure it
ReplyDeleteAh, no Homer, the explanation is that I am eminently qualified and know what I am talking about - and you don't.
ReplyDeleteYou cannot even tell me why you think such a question cannot be answered, let alone criticise an intervention analysis method of investigation.
Still no answer from Homer.
ReplyDeleteWhat a squib. We can conclude that the pink batts programme made houses less safer.
Mark,
ReplyDeletebeing some-one who doesn't understand basic statistics you are mixing the number of houses that had fires with the fact how long after installation fires started.
I do understand basic statistics Homer. As well as more advanced time series and microeconometrics.
ReplyDeleteYou don't understand any of this.
I gave you a reference, now please go and consult it and learn the material before I ask you again solemnly to retire.
"Mixing the data up"
What a pathetic excuse. It is the same time series dataset with a structural break.
The frequency of fires is simply an attribute of the variable(s).
a structural break. really then you do not know even basic statistics. hint we do not have enough observations to say that.
DeleteMark there were fires.
We know there were about ten times less fires in the insulation scheme.
You are trying to say how early the fires were is more important.
they are entirely different and by the way it is entirely consistent with what I have written
"a structural break. really then you do not know even basic statistics. hint we do not have enough observations to say that."
ReplyDeleteProve it. Where is your power test? We can have a panel on a state by state basis but you'd have to fix the effects for different state based regulation. But we'd have our dataset.
You see 'structural break' and simply conclude we don't have enough data points, you refuse to do anything further than basic analysis, break down the time series into componenets or do an intervention analysis.
You are way out of your depth Homer. What you learn about econometrics in an MBA back in 1985 would have been utterly pathetic.
"We know there were about ten times less fires in the insulation scheme."
So forty people died during the time from other roofing cavity installations did they?
"You are trying to say how early the fires were is more important."
Yes it is, it is a time series.
Just admit it failed as a GDP enhancing policy and killed people and burned houses down. Stop making excuses for Rudd and Gillard.
Mark,
ReplyDeleteI am very happy for you to produce evidence of this. however you never have!
Just for the record I have studied staistics at four universities and econometrics both at an undergraduate and postgraduate level in economics.
I don't have to anything other than a basic Z test to show there is a significant difference between the two averages.
A structural break on the other hand does need a fair amount of data!
"Just for the record I have studied staistics at four universities and econometrics both at an undergraduate and postgraduate level in economics"
ReplyDeletePlease name your qualifications.
"I don't have to anything other than a basic Z test to show there is a significant difference between the two averages."
You have shown the pink batts more houses more dangerous and you don't even realise it.
A structural break is not the only time series method available. properly speaking, it is not a 'time series method'.
You are a dangerous idiot, Homer.
Please retire.
Mark,
ReplyDeleteyou have never once got past all of the jargon.
the reason is you have no idea.
Unless you can produce your 'structural break' your comments will go the same way
You are a coward, Homer.
ReplyDeleteYou have just been bested and don't want the evidence to stay around.
You don't even know what I am talking about now Homer. You keep on referring to a structural break like it is a gold standard of proof.
It isn't.
You don't even understand you have shown there is a structural break - nevertheless you are assuming that means being different implies a falling trend in the second portion of the sample.
You are an incompetent fool.
Please retire.
there are two observations and you are saying thee is a structral break.enough said.
ReplyDeleteyou can make ignorant stament at cultallaxy but not here. either show it or go away.
"there are two observations and you are saying thee is a structral break.enough said."
ReplyDeleteNo, you are lying.
1. Monthly data exists (which that incompetent "Possum" used), and I am saying look at it as a panel.
2. I am also saying to use difference equation based time series methods - which you are unfamiliar with and you refuse to familiarise yourself with despite my best efforts to generously educate you.
Crawl down your hole and keep on defending this shambles of a Government you "are no longer prepared to defend", Homer.
Actually, what I really want from you is a promise that you will never work in the education system again. It would be a public service.
Really I read all the relevant reports and they had only yearly data and they were not precise.
ReplyDeleteyou are undoubtedly mixing things again.
you have showed on previous occasions you do not know how many observations are needed to make the statement that something is statistically significant!
"Really I read all the relevant reports and they had only yearly data and they were not precise."
ReplyDeleteYou are being elusive and making up solid conclusions with your own poor analysis and sandbagging anything done to the contrary.
"you are undoubtedly mixing things again."
No I am not Homer. This is beyond your ken. Despite you studying statistics at 4 unis.
"you have showed on previous occasions you do not know how many observations are needed to make the statement that something is statistically significant!"
It's not 30 or 200 Homer. This is what lazy, sloppy "thinkers" like you believe. It can be deduced with a power test if something has enough statistical significance.
You are out of your depth.
The policy was a macroeconomic and safety failure.
You are still shilling for this shambles of a Government.
Please retire, before you ruin the minds of our young people.
Mark,
ReplyDeletei am getting very bored with stating something but unable to back it up.
I am more than happy for you to do that however I am not happy for you to come over here and lambast people with your baselss points.
Homer were my points were not baseless. You simply have no rebuttal to them.
ReplyDeleteMark, last time.
ReplyDeleteyou claimed the marginal effect got worse.NO evidence
you claimed a structural break, no evidence.
As I said I am more than happy for you to present an argument but i have lost patience with your aspergers claims.
No Homer, you are simply claiming a trend reparameterises to whatever the hell you and possum prejudge it to be, then you ignore any other possible method of analysis.
ReplyDeleteYou are trying to prove that something is true by not falsifying it but by ignoring alternative analyses, and reshaping parameters as you see fit.
There is absolutely no proof that homes not in the pink batt scheme were made safer. There however is an impulse function for fires in this scheme. You would understand that if you followed the reference I gave you to Enders.
Please retire.
Last chance Mark,
ReplyDeleteeither you put up or I will shut you up.
Homer
ReplyDeleteI have attacked your position and you have been found wanting.
You will not admit the pink batts were an economic, environmental and safety failure and you cannot argue the issue because you lack competency - both as an economist and in statistic analysis. You still also support the same Government responsible for the disaster.
Please retire.
Mark, you always make a lot of claims but never produce any evidence to support your position.
ReplyDeleteYou have again been found wanting. Unless you wish to produce any evidence to support your theories this is it.