「華人戴明學院」是戴明哲學的學習共同體 ,致力於淵博型智識系統的研究、推廣和運用。 The purpose of this blog is to advance the ideas and ideals of W. Edwards Deming.

2016年7月31日 星期日

Bill Gates Views Good Data as Key to Global Health. Error bars

Let’s talk about the Global Burden of Disease study. [The GBD is published on a nearly annual basis by the IHME.] Because this study is an independent and now—with the additional funding your foundation is providing—a regularly updated assessment of global health, it could in principle serve as a gold-standard reference of progress in various parts of the world on various diseases. Have you actually used it in that way to identify programs that are working and those that are not working and need redirection?
The GBD assembles data from lots of different field studies, many of which we are funding. For example, we ran a big study called GEMS[Global Enteric Multicenter Study] to try to figure out all the different causes of diarrheal disease—rotavirus is the biggest cause, but there’s alsoE. coli, shigella, cryptosporidium and others—and how important each one is. We still struggle with large uncertainty about the locations and extent of certain diseases, such as typhoid and cholera, which no country wants to admit they still have. My teams, like others who are very active in fieldwork, usually are looking at the primary papers as soon as they come out in the scientific literature. By the time the information gets aggregated and vetted and incorporated into the GBD database, it should no longer be surprising to us.
But GBD is super helpful when we’re talking to developing countries and saying “Look, here’s what is going on with tuberculosis in your country versus others like yours.” It’s a very important tool to educate people—like the World Development Report was for me. You can see the time progressions and zoom in on any country. It’s one of the better data-visualization sites in the entire Web. It’s super nice. And most people aren’t that up to date on these disease trends—particularly for infectious diseases. So I’ve been taking GBD charts with me when I’ve met with people in Cambodia or Indonesia or even at the French aid agency about trends in francophone Africa. They can reveal when we haven’t set the right priorities—so it’s a very important tool for me. Before I go into strategy meetings, I sometimes look at the GBD to remind myself of the numbers.
We now have enough detailed data to break big illness categories like diarrheal disease apart into separate diseases by the root cause. Even so, the error bars tend to be quite large on these estimates because, unlike in the rich world where disease cases are actually counted and tracked, in the poorer parts of the world we have to rely on sampling and extrapolation. If you happen to sample in places where the condition is unusually prevalent, the extrapolated numbers can be wrong.
That raises an interesting point. One of the potential advantages of having this statistical inference machine that IHME uses to produce the GBD estimates is that you could identify where you would get the most bang for the buck if you did a new study that will improve the empirical input to the system. Has the GBD actually been used to prioritize funding of surveillance in this way?
Oh, yeah. Disease surveillance in the poor world is terrible. While it’s great that we now have this published set of numbers, they have pretty big error bars—and probably some of the error bars should be even bigger than they are shown in the IHME reports. But we’re actively looking at ways to improve the situation. New diagnostics are becoming available, for example, that can check for lots of different diseases by analyzing just a few drops of blood. So rather than running one study after another, each of which has to set up a bunch of different centers just to get data on one kind of disease, we might be able to use clinics that are running all the time and constantly monitoring the prevalence of lots of diseases simultaneously.

From Wikipedia, the free encyclopedia
bar chart with confidence intervals (shown as red lines)
Error bars are a graphical representation of the variability of data and are used on graphs to indicate the error, or uncertainty in a reported measurement. They give a general idea of how precise a measurement is, or conversely, how far from the reported value the true (error free) value might be. Error bars often represent one standard deviation of uncertainty, one standard error, or a certain confidence interval (e.g., a 95% interval). These quantities are not the same and so the measure selected should be stated explicitly in the graph or supporting text.
Error bars can be used to compare visually two quantities if various other conditions hold. This can determine whether differences are statistically significant. Error bars can also suggestgoodness of fit of a given function, i.e., how well the function describes the data. Scientific papers in the experimental sciences are expected to include error bars on all graphs, though the practice differs somewhat between sciences, and each journal will have its own house style. It has also been shown that error bars can be used as a direct manipulation interface for controlling probabilistic algorithms for approximate computation.[1] Error bars can also be expressed in a plus-minus sign (±), plus the upper limit of the error and minus the lower limit of the error.[2]

Bill Gates has a well-established knack for sifting through complex data sets to find the right pathways for making progress around the globe in health, education and economic development.
In an interview with Scientific American the philanthropist talks about the statistics that inspire him most

2016年7月27日 星期三








2016年7月24日 星期日

“The Market for Lemons”, lemon law, Phishing for Phools: The Economics of Manipulation and Deception.

Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism

2016年7月20日 星期三

會腐蝕的鋁製 Apple Watch

First indication that galvanic corrosion is a potential problem ...

Many here are planning to get the aluminum Sport model and one of the bands from the stainless steel collection to wear to nicer occasions. However, all the non-Sport bands (except the Leather Loop) house stainless steel connectors, and we know that aluminum and stainless steel together poses a bi-metallic corrosion risk, also known as galvanic corrosion, especially in the presence of sweat (from working out or just from wearing the watch on a hot day).

從2010.03寫到現在,我只是想寫 --…

2016年7月19日 星期二

VW Emissions Cheating Ran Deep and Wide, State Alleges (wsj)

VW Emissions Cheating Ran Deep and Wide, State Alleges
Deception went beyond ‘couple of software engineers,’ New York attorney general says


Volkswagen AG’s emissions cheating spanned more than a decade and arose from deliberate efforts by dozens of employees to mislead regulators and consumers about diesel-powered vehicles, according to a lawsuit from New York’s top law-enforcement official.
The decision to use software to manipulate emissions tests traced back as far as 1999, when engineers at the company’s Audi luxury unit developed technology to quiet diesel vehicles, according to a lawsuit filed Tuesday by New York Attorney General Eric Schneiderman. The technology, rolled out in 2004, made vehicles exceed European emissions standards, so the engineers added software they called “acoustic function” to turn it off during emissions tests, the lawsuit said.
Volkswagen eventually developed the software, known as defeat devices, for diesel vehicles sold in the U.S. starting in 2008. The German auto maker in September admitted to U.S. environmental regulators that the vehicles used software that allowed them to pollute on the road up to 40 times the amount allowed. The Environmental Protection Agency’s subsequent disclosure of the cheating sparked probes across the globe and forced the resignation of Chief Executive Martin Winterkorn.
The latest lawsuit cites emails and other documents to allege a prolonged effort among dozens of Volkswagen employees in the U.S. and Germany to equip vehicles with the devices and stonewall inquiries from regulators. The deception went far beyond the “couple of software engineers” whom Michael Horn, then Volkswagen’s top U.S. executive, blamed in October, according to the lawsuit. Some managers, engineers and executives named in the suit haven’t been previously identified; others have been suspended or resigned since regulators disclosed the cheating.
In June, Volkswagen agreed to pay up to $15 billion to settle claims with environmental regulators, owners of 475,000 vehicles with two-liter diesel engines and some state authorities. The company still faces legal claims affecting more than 80,000 three-liter diesel vehicles and a U.S. Justice Department criminal probe. The software is on some 11 million vehicles world-wide, Volkswagen has said.
It is regrettable that some states have decided to sue for environmental claims now... 
—Volkswagen spokeswoman
Mr. Schneiderman’s suit seeks up to $450 million in civil penalties for what it calls Volkswagen’s “egregious and pervasive violations” that “strike at the heart” of state environmental laws, and were “the result of a willful and systematic scheme of cheating by dozens of employees at all levels of the company.” Massachusetts and Maryland filed similar lawsuits on Tuesday.
A Volkswagen spokeswoman called Tuesday’s allegations “essentially not new,” adding the company has been addressing them in discussions with U.S. and state authorities. “It is regrettable that some states have decided to sue for environmental claims now, notwithstanding their prior support of this ongoing federal-state collaborative process,” she said. Volkswagen declined interviews with those individuals mentioned in the New York suit.
In 2006, Volkswagen engineer James Liang developed a defeat device for a Jetta diesel car and eight years later conducted tests to help conceal emissions-cheating, according to the lawsuit. Oliver Schmidt, who headed Volkswagen’s U.S. regulatory compliance office in 2014 and early 2015 also played a key role in the deception, the lawsuit alleged.


0:00 / 0:00

Volkswagen has settled emissions claims with regulators and owners of about a half million diesel-powered vehicles. The settlement terms were announced Tuesday by environmental regulators. The WSJ's Lee Hawkins explains.
A 2007 Volkswagen memo recapping a meeting with California regulators said a state official “expects emission-control systems to work during conditions outside of the emissions tests. Volkswagen agrees,” according to the lawsuit. Instead, Volkswagen engineers and executives developed and implemented emissions-increasing defeat devices “as part of the normal course of business,” the lawsuit alleged.
In May 2014, a senior Volkswagen executive warned then-CEO Mr. Winterkorn of growing suspicions from regulators and that the company couldn’t explain dramatic emissions increases, the lawsuit said. A year later, a manager admonished a Volkswagen official in the U.S. for allowing another employee to send a frank email expressing concerns, it added.
The emissions-cheating scandal began in earnest in late 2006, when Volkswagen, facing engineering challenges, adapted technology Audi had developed to address other emissions problems and installed the defeat-device software on hundreds of thousands of Jetta, Golf and other cars, the lawsuit alleged. Cars with model years 2009 through 2015 ended up with the software.
In October 2006, many Volkswagen executives held a conference call with California regulators, with the latter requesting additional details on emissions-control devices.
Do we need to discuss next steps? Come up with the story please!
—VW executive’s email
Later that year, Leonard Kata, a Volkswagen emissions regulations and certification manager, emailed colleagues that government officials were interested in whether emissions-control devices were illegal defeat devices and detailed how agencies make such determinations, said the lawsuit.
Volkswagen executives in subsequent years discussed the development and use of defeat devices to dupe emissions tests, including a direct report to Mr. Winterkorn, the CEO; heads of Audi’s powertrain development; and other Volkswagen division heads and employees below them, the lawsuit alleged.
After learning in spring 2015 that California regulators planned to conduct further tests on vehicles, internal Volkswagen emails “began to reflect desperation and panic,” the lawsuit said. One engineering executive conveyed “serious concern regarding” what California’s tests “would expose,” the lawsuit alleged. “Do we need to discuss next steps?” the engineering executive wrote, according to the lawsuit. “Come up with the story please!”
A senior emissions-compliance manager in the U.S. for Volkswagen, meanwhile, expressed concern in a May 2015 email that requests for information from California regulators were “not a normal process and that we are concerned that there may be possible future problems/risks involved.”
The email added that Volkswagen’s software was being reviewed by top officials at California’s environmental agency, according to the lawsuit.

"Clean energy won’t save us – only a new economic system can do that"

"What would we do with 100% clean energy? Exactly what we are doing with fossil fuels: raze more forests, build more meat farms, expand industrial agriculture, produce more cement, and fill more landfill sites, all of which will pump deadly amounts of greenhouse gas into the air. We will do these things because our economic system demands endless compound growth, and for some reason we have not thought to question this."
It’s time to pour our creative energies into imagining a new global economy. Infinite growth is a dangerous illusion

2016年7月17日 星期日

3,000 studies using brain scans could simply be wrong;法國內政部:"SAIP應用程序過晚發送了有關7月14日尼斯恐怖襲擊的消息"

Software glitches blow a hole in a lot of neuroscience
Two studies, one on neuroscience and one on palaeoclimatology, cast doubt on established results. First, neuroscience and the reliability of brain scanning

而當週四(7月14日)夜間 尼斯發生恐怖襲擊後,這款應用在當地時間15日凌晨1:34才發出第一條訊息,晚了近三個小時。法國國慶日當晚,在南部海濱城市尼斯,一名31歲的突尼斯裔法國公民駕駛大型載重車衝入看煙火的人群,導致至少84人死亡,上百人受傷。
法國迴聲報(Les Échos)援引政府消息人士報導稱,當地政府準備好的訊息應該於23:15發出,然而這款法國公司Deveryware設計的應用軟件卻因技術故障而未能及時完成任務。

the Affordable Care,Journal of the American Medical Association;試驗登記不可或缺

A famous Columbia graduate just published an article in The Journal of the American Medical Association.

Paper in a Top Medical Journal Has an Unexpected Author
The JAMA paper highlights some of the successes of the Affordable Care Act

The JAMA paper highlights some of the successes of the Affordable Care Act

By Rachael Rettner, LiveScience on July 14, 2016It may be the first time a sitting president has authored a complete academic article — with an abstract, findings and conclusions — that's been published in a scientific journal, at least in recent history. Credit: Official White House Photo by Lawrence Jackson

In an unusual move for a sitting president, Barack Obama has published a scholarly paper in a scientific journal.

The paper, which discusses the success and future of the Affordable Care Act (ACA), was published Monday (July 11)in the prestigious medical journal JAMA.

It may be the first time a sitting president has authored a complete academic article — with an abstract, findings and conclusions — that's been published in a scientific journal, at least in recent history. However, several other presidents have written commentaries or opinion pieces that have been published in scientific journals during their presidency, including George W. Bush, who wrote about access to health care in a paper published in JAMA in 2004, and Bill Clinton, who wrote a commentary published in the journal Science in 1997.

Obama's journal article analyzes data gathered from other reports and studies, and highlights some of the successes of the ACA, including a drop in the percentage of Americans who do not have health insurance . After the act became law, the uninsured rate dropped by 43 percent, from 16 percent of Americans in 2010 to 9.1 percent in 2015, the paper says. [The 5 Strangest Presidential Elections in US History]

Still, Obama said, the country continues to face challenges on the way to improving its health care system. "Despite this progress, too many Americans still strain to pay for their physician visits and prescriptions, cover their deductibles, or pay their monthly insurance bills; struggle to navigate a complex, sometimes bewildering system; and remain uninsured ," Obama wrote.

To make sure Americans have enough insurance options and to keep insurance costs low, Obama encouraged Congress to revisit the "public option" plan, meaning a government-run insurance plan that would compete in the insurance marketplace alongside private plans. This public option could be available in parts of the country where insurance options are limited, he said.

Obama also recommended policies that could help reduce the cost of prescription drugs, including those that "give the federal government the authority to negotiate prices for certain high-priced drugs."

Obama's article was not peer-reviewed, but it went through several rounds of editing and fact-checking, according to Bloomberg.
Top 10 Ailing Presidents
Obamacare, Nixoncare: Health Care Debates Are All About Politics
7 Facts You Should Know About Health Care Reform

Copyright 2016 LiveScience, a Purch company. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.

試驗登記不可或缺 2012英國《金融時報》專欄作家蒂姆•哈福德
我剛剛搖了幾次六面骰子,搖出的結果是:6、5、5、6、5。我的問題是:你覺得這個骰子有偏向性嗎?一種思路是,考慮在純隨機情況下只搖出5和6的概率。概率不大:242分之1。在你說我對骰子做了手腳之前,請先讓我說一些剛才忘記提到的事情:除了5和6之外,我還搖出了一次4、三次3、兩次2和一次1。但我對這些結果不感興趣,所以沒告訴你。如果你認為被忽略的結果也具有意義的話,這意味著你開始認識到“試驗登記”(trial register)的重要性——儘管這個詞聽上去有點書呆子。


我報告了偏向大數字的搖骰子結果。你或許會不客氣地懷疑,在行業贊助的藥物臨床試驗中,顯示藥物良好效果的試驗更有可能公之於眾。你的懷疑是對的——本•高達可(Ben Goldacre)在新書《壞藥商》(Bad Pharma)中精闢地總結了這種現象。以混亂或災難告終的試驗有可能“失踪”:它​​們的價值也許夠不上發表學術論文,但結果必須記錄下來。如果試驗結果過於枯燥,令研究人員寫不出符合發表標準的論文,同樣會導致試驗“失踪”。


幾年前,國際醫學雜誌編輯委員會(International Committee of Medical Journal Editors)宣布,在其控制下的知名期刊,將不再發表以臨床試驗為基礎、但事先未對試驗進行登記的研究。此舉效果驚人,顯著提高了新試驗的登記率。但不幸的是,西爾萬•馬蒂厄(Sylvain Mathieu)等人2009年在《美國醫學協會雜誌》(Journal of the American Medical Association)上解釋道,在他們檢查的研究中有超過半數無視了這一規則,卻仍然成功發表。未登記就不發表的威脅似乎是空洞的。

那麼,如何對待近年來飽嚐隨機試驗甜頭的經濟學?好消息是,美國經濟學會(American Economic Association)正要為經濟學試驗創立登記制度,將於明年實施。登記目前是自願的,但兩位領先經濟學家——麻省理工學院(MIT)的埃斯特•迪弗洛(Esther Duflo)和耶魯大學(Yale)的迪恩•卡蘭(Dean Karlan) ——對我表示,他們認為很有希望形成強大的社會規範,支持試驗登記。



作者是英國廣播公司廣播四台(BBC Radio 4)《或多或少》(More or Less)節目的主持人譯者/劉鑫