Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : Booms, Busts, and Recoveries -- Ignore unavailable to you. Want to Upgrade?

To: GraceZ who wrote (39444)	10/10/2003 2:51:33 PM
From: Joe S Pack	Respond to of 74559

The question is not whether statistics is useful or not, nor sampling is valid or not. It has been used in all places from manufacturing to census. But these two quote are equally true as well: "Lies! Damn Lies!" . (There are three kinds of lies: lies, damn lies and statistics. Mark Twain (1835-1910)) "Garbage in Garbage out". You can manipulate any model in any way you want. It all depends on how you collect data, how you sanitize, and how you interpret the result. Manipulation is universal as long as they can get away. Here is another example of government manipulation and is bringing politicians into action in UK. news.independent.co.uk Statistics chief faces double inquiry into errors By Philip Thornton, Economics Correspondent 10 October 2003 The UK's chief statistician has been summoned to appear before MPs next week as it emerged yesterday that the Government's watchdog is launching an investigation into the quality of official figures. News of the inquiries comes in the wake of a series of dramatic revisions to key economic data that prompted a debate over the wisdom of the Bank's decision to cut rates in July. Len Cook, the National Statistician, faces a rough ride from the Commons' Treasury Select Sub-Committee on Wednesday. Yesterday he told the Independent: "I can defend the competence of my department in any environment. "We make the odd mistake and we've made more mistakes that I expected us to do but we produce a huge variety of statistics and we get a large proportion of them right first time." He welcomed the review and said that, as a New Zealander, he had been surprised by the scale of criticism In this day and age, there is absolutely no reason why, especially the jobless claims, cannot be actual to the last guy who walked in to the unemployment office Friday pm in Hawaii (that being the latest time zone for US). Many unemployed don't ever show up in an unemployment office. They do know the exact number of jobless claims (within a small margin of error) but this is not what defines unemployment. To understand why this is so all you have to do is ask yourself how it is that you count someone in the ranks of the unemployed, whose unemployment benefits have run out (they last six months) but is still looking for a job. Under the current definition of unemployment, this person is unemployed but would not come up in the stats for unemployment insurance claims. Using statistical sampling that person is counted. The unemployment rate differs significantly from those receiving unemployment insurance during times when jobs are harder to find. Also, how about self employed people whose business has failed who are looking for a job, they don't get unemployment insurance payments. How about a woman whose children are grown up and is looking for work? The list goes on and on of these kinds of situations. Statistical sampling is a very accu

To: GraceZ who wrote (39444)	10/10/2003 3:00:53 PM
From: Ramsey Su	Respond to of 74559

Grace, Here's a pretty good layman's explanation of the methods employed and the reasoning behind them. If you want to argue the validity of their model on a scientific basis you have to order the white paper detailing the method precisely, but I'll warn you, you better have a background in statistics to get through it: Thanks for your advice. Let me try and if I have any questions, I will for sure ask. By the way, I have read some of your posts on the real estate thread. They are great samples of your vast knowledge in that field also. Ramsey

To: GraceZ who wrote (39444)	10/10/2003 5:51:57 PM
From: smolejv@gmx.net	Read Replies (1) \| Respond to of 74559

Hi Grace: Can one sum it up like this: the population of those coming up with jobless claims does not mirror the population of the jobless. iow equating the two is misleading and/or wrong because of biased sampling. RegZ dj

To: GraceZ who wrote (39444)	10/10/2003 6:37:05 PM
From: maceng2	Read Replies (1) \| Respond to of 74559

Hi Grace, I would regard someone who is "unemployed" as someone who is looking for work or who is collecting unemployment benefit. They have a social security number and it's fairly easy to find out if they are getting a paycheck or not. So why the need for sampling? We have lots of computers to process the data. OK there is no direct measurement for the black economy (unrecorded work and pay, unrecorded crime etc), but some models can be made for that and some statistical measurements made to guestimate it's influence. My point is the real unemployment number can be found with a little bit of effort and imagination. It's not as if we don't have the technology.