Ethnic unrest has been a common theme in China recently (mysteriously for many, but not for me). One source of that comes from Tibet. I was once asked why Chinese government treats Dalai Lama with such hostility and distrust? Why would China distrust such a peace-loving, caring, charismatic and perfect Nobel Peace prize winner? Digging a little bit into history, with basic information from wikipedia, I find the distrust makes perfect sense.
So a brief history is in order (from wiki): On 17 November 1950, at the age of 15, the 14th Dalai Lama (the one you know) was enthroned formally as the temporal ruler of Tibet. He sent a delegation to Beijing, which ratified the Seventeen Point Agreement for the Peaceful Liberation of Tibet. (which allowed a high degree of autonomy) He worked with the Chinese government: in September 1954, together with the 10th Panchen Lama he went to the Chinese capital to meet Mao Zedong and attend the first session of the National People's Congress as a delegate, primarily discussing China's constitution. On 27 September 1954, the Dalai Lama was selected as a deputy chairman of the Standing Committee of the National People's Congress, a post he officially held until 1964.
However, Dalai Lama was planning something else. He accepted assistance from CIA. In 1956, a large rebellion broke out in eastern Kham, an ethnically Tibetan region in Sichuan
province. To support the rebels, the CIA launched a covert action
campaign against the Communist Chinese. A secret military training camp
for the Khampa guerrillas was established at Camp Hale near Leadville, Colorado, in the U.S. The guerrillas attacked Communist forces in Amdo and Kham but were gradually pushed into Central Tibet.
At the outset of the 1959 Tibetan uprising, fearing for his life, the Dalai Lama and his retinue fled Tibet with the help of the CIA's Special Activities Division, crossing into India on 30 March 1959, reaching Tezpur in Assam on 18 April.
This excerpt from wikipedia should have explained the origin of the distrust. Oran's Dictionary of the Law (1983) defines treason as "...[a]...citizen's actions to help a foreign government overthrow, make war against, or seriously injure the [parent nation]."(from wiki). I think what Dalai Lama did fit the definition of treason. Then it is not all that surprising that China treats a traitor with suspicion, distrust, and hostility, especially after that traitor has publicly embarrassed China many times afterwards. To build trust is a lengthy process, and once destroyed, it is hard to rebuild, isn't it?
While Dalai Lama likes to instill some Tibetan teaching, I wish to quote Conscious: I used to judge a person by what he says, now I have learned to judge a person by what he does.
Sunday, August 5, 2012
Should race be a factor in statistical inference?
This is a question I have long been thinking about. Ye's incident brought this question back to my attention. While I despise and lament the racist comments concerning Ye's performance, I should give a qualified "yes" to the question posed in the title. While apparently paradoxical, this response does not contradict my attitudes on Ye's matter, as I pointed out in the previous post that race should be accounted for properly in a scientific manner.
Conditioning is the soul of statistics. Whenever possible, we should condition on relevant information to aid in our inference. Race, in some cases, could be a relevant information. When that is the case, we should not simply discard the information for political correctness. Being fair-minded does not mean we should turn our head away from any information concerning gender and race. We are seeking truth, uncomfortable it might be, and when those information contains information about the truth, we choose to extract the information from it. (The book Intelligence Paradox contains a good discussion relating to academic political correctness in its Introduction, which I wish to quote in my incoming blogs, and which I agree whole-heartedly.) The hallmark of a racist is not reading information from race, but rather read non-existent information from race and refuse to consider other information.
However, what is lamentable is that many people do not do it properly. We might have different priors based on race, but those priors cannot be dogmatic. They should contain information about race, but they should recognize the limited scope of the information, and allow enough flexibility in the prior so as to readily accept any new relevant information. Put in another way, even if we allow the priors to be different, this difference will be dominated once new relevant information becomes available. Thus, incorporating race, in most circumstances, will only be pertinent in theory, and non-significant in practice, unless one has very strong evidence that the racial information is extremely relevant and reliable, a case for which I am having an extremely difficult time to construct an example.
The problem with most racial information people wish to incorporate is that they are not "structural"--- The relation between race and certain behavior are not constant or fundamental. Given that most of the information is historical data, and that the possibility of a structural change, the past observations contains little relevant information for today, and it will be detrimental to make decisions based on past non-structural observations (similar to the Lucas Critique in economics). For example, African Americans might have lower literacy historically, but this association is non-structural, and it will (and did) change as time passes by. In general, blindly incorporating historical information as if they are historical is a sign of sloppy thinking and crappy reasoning, both of which might bring severe consequences. The current financial crisis, as some economists would argue, results from people's extrapolating from historical data on real estate (believing future performance will mirror past performance). Those traders who were stupid enough to commit such fallacy, brought tremendous loss to their companies. In the case of Ye, the Chinese swimming team had a dishonorable history in the 1990's, and that was the past. To blindly project 1990's to 2010's is similar to what those garbage traders did, except in this case, the consequences is born by the potentially innocent Chinese swimmer Ye.
A less innocuous phenomenon is confirmation bias. When the information of race is entered, the person directly forms a (racist) belief, and he looks for information that only confirms his belief, and interpret ambiguous information so as to confirm his belief. In the case of Ye, the history is just information he digs up to rationalize his belief, he discard relevant information (other swimmers could improve more significantly from personal best and drug test), and he biasedly interprets ambiguous information (interpreting Ye's second not-as-startling performance as evidence of backing-off, rather than symptom of fatigue after staying up late for drug tests nights in a row till right before the contest and extreme psychological pressure resulting from media defamatory coverage).
To conclude, while incorporating racial information per se is scientific and non-racist, many tend to err on the side of over-incorporating racial information. The real racist behavior is not just to incorporate such information, but to do it improperly. The real danger results from forming a dogmatic prior and indulging in confirmation bias.
Conditioning is the soul of statistics. Whenever possible, we should condition on relevant information to aid in our inference. Race, in some cases, could be a relevant information. When that is the case, we should not simply discard the information for political correctness. Being fair-minded does not mean we should turn our head away from any information concerning gender and race. We are seeking truth, uncomfortable it might be, and when those information contains information about the truth, we choose to extract the information from it. (The book Intelligence Paradox contains a good discussion relating to academic political correctness in its Introduction, which I wish to quote in my incoming blogs, and which I agree whole-heartedly.) The hallmark of a racist is not reading information from race, but rather read non-existent information from race and refuse to consider other information.
However, what is lamentable is that many people do not do it properly. We might have different priors based on race, but those priors cannot be dogmatic. They should contain information about race, but they should recognize the limited scope of the information, and allow enough flexibility in the prior so as to readily accept any new relevant information. Put in another way, even if we allow the priors to be different, this difference will be dominated once new relevant information becomes available. Thus, incorporating race, in most circumstances, will only be pertinent in theory, and non-significant in practice, unless one has very strong evidence that the racial information is extremely relevant and reliable, a case for which I am having an extremely difficult time to construct an example.
The problem with most racial information people wish to incorporate is that they are not "structural"--- The relation between race and certain behavior are not constant or fundamental. Given that most of the information is historical data, and that the possibility of a structural change, the past observations contains little relevant information for today, and it will be detrimental to make decisions based on past non-structural observations (similar to the Lucas Critique in economics). For example, African Americans might have lower literacy historically, but this association is non-structural, and it will (and did) change as time passes by. In general, blindly incorporating historical information as if they are historical is a sign of sloppy thinking and crappy reasoning, both of which might bring severe consequences. The current financial crisis, as some economists would argue, results from people's extrapolating from historical data on real estate (believing future performance will mirror past performance). Those traders who were stupid enough to commit such fallacy, brought tremendous loss to their companies. In the case of Ye, the Chinese swimming team had a dishonorable history in the 1990's, and that was the past. To blindly project 1990's to 2010's is similar to what those garbage traders did, except in this case, the consequences is born by the potentially innocent Chinese swimmer Ye.
A less innocuous phenomenon is confirmation bias. When the information of race is entered, the person directly forms a (racist) belief, and he looks for information that only confirms his belief, and interpret ambiguous information so as to confirm his belief. In the case of Ye, the history is just information he digs up to rationalize his belief, he discard relevant information (other swimmers could improve more significantly from personal best and drug test), and he biasedly interprets ambiguous information (interpreting Ye's second not-as-startling performance as evidence of backing-off, rather than symptom of fatigue after staying up late for drug tests nights in a row till right before the contest and extreme psychological pressure resulting from media defamatory coverage).
To conclude, while incorporating racial information per se is scientific and non-racist, many tend to err on the side of over-incorporating racial information. The real racist behavior is not just to incorporate such information, but to do it improperly. The real danger results from forming a dogmatic prior and indulging in confirmation bias.
Saturday, August 4, 2012
Another on Ye
First, let me quote this article.
http://www.nature.com/news/why-great-olympic-feats-raise-suspicions-1.11109
This article, once again attempt to justify accusations of doping on Ye with two rationals:
1) such a fast swim is an anomaly. 2) Drug test cannot rule out the possibility of doping.
These two might sound convincing and reasonable justifications, and let one to conclude that after all, all the accusations are not racism in nature, but based on hard scientific reasoning. Unfortunately, this is only wishful thinking, and the facade of scientific objectivity cannot withstand close scrutiny. To justify these accusations from these two observations, requires committing some serious fallacies in statistical reasoning.
First, there is a difference between how likely one is able to swim that fast without doping and how likely has she not been involved in doping given that she swam so fast. The low likelihood of the former does not necessarily imply the low likelihood of the latter. This is called Prosecutor's Fallacy. (See http://en.wikipedia.org/wiki/Prosecutor%27s_fallacy) According to Wikipedia (I know), "deliberate use of the prosecutor's fallacy is prosecutorial misconduct and can subject the prosecutor to official reprimand, disbarment or criminal punishment". Hence it is not some arcane fallacy I dig up to defend Ye. It is disappointing to witness that a magazine like Nature would allow such floppy reasoning. What is of concern, is conditional on such performance, how likely has Ye been involved in doping. The article, and the mass media, fails to give hard data to justify their accusation.
Second, on a related note, one friend on facebook pointed out: "What's interesting is that Katie Ledecky, a 15 year old, won the 800 free today. She shaved 21 second off her best time from last year. Ye Shiwen shaved 5 seconds off her best from last year (in a 400 meter race). So perhaps what Ye Shiwen did isn't that surprising." If Katie can improve 21 seconds without invoking any suspicion, it is very hard to justify accusations on Ye without resorting to racial factor (which could be justified, and will be the topic of my next blog). I am OK with taking racial factors into account, as long as 1) it is done properly (not leading to dogmatic prior) 2) it is openly acknowledged instead of denied.
Finally, the argument that test is not conclusive is red herring. The article seems to hint that now that the drug test does not rule out the possibility of Ye's doping, we should not drop our suspicion in response to the drug test result. However, every statistical test is not conclusive, and only suggestive. The key is that conditional on the result of the test, we do change our belief! At least, the likelihood of Ye doping is significantly reduced given the test result. The power of the test might not be high (which I think the article correctly point out), and we, as a result, will not as dramatically update our belief as what we would do if the power of the test is high, but to not to update our belief at all given the test result is to choose to deliberately ignore relevant evidence and information, not the trademark of a rational being. Consider this, SAT is a test for intelligence, though the power of the test is low, that is people might get higher score even if they are of mediocre intelligence. However, when you first encounter someone, who might look stupid, but you learnt that he got a really high score on SAT, you will need to update your belief about his intelligence if you are a rational being. Indeed, if you have little prior information about his intelligence (say the appearance is not a good indicator of intelligence), you will have to update your belief dramatically, despite the low power of SAT. This might sound paradoxical, but it reflects sound reasoning---in the absence of relevant information, a noisy signal, when compared to other information (that is null, and is of infinite noise), the relative noise ratio is zero! Unsurprisingly, you put all weights on this information (this is the idea behind Fisher Weighting). Hence, unless you have a pretty dogmatic prior concerning Ye, when the drug test is the only available relevant information, one would need to update his belief significantly, regardless the alleged power of the test.
These two logical flaws might be unnoticeable under first reading, but jumps out for any reader with a sense of basic (but correct) intuition about probability. It is a pity that Nature calls it fair-minded, but what is really annoying is the absence of sound logic under the pretense of sound logic.
http://www.nature.com/news/
This article, once again attempt to justify accusations of doping on Ye with two rationals:
1) such a fast swim is an anomaly. 2) Drug test cannot rule out the possibility of doping.
These two might sound convincing and reasonable justifications, and let one to conclude that after all, all the accusations are not racism in nature, but based on hard scientific reasoning. Unfortunately, this is only wishful thinking, and the facade of scientific objectivity cannot withstand close scrutiny. To justify these accusations from these two observations, requires committing some serious fallacies in statistical reasoning.
First, there is a difference between how likely one is able to swim that fast without doping and how likely has she not been involved in doping given that she swam so fast. The low likelihood of the former does not necessarily imply the low likelihood of the latter. This is called Prosecutor's Fallacy. (See http://en.wikipedia.org/wiki/Prosecutor%27s_fallacy) According to Wikipedia (I know), "deliberate use of the prosecutor's fallacy is prosecutorial misconduct and can subject the prosecutor to official reprimand, disbarment or criminal punishment". Hence it is not some arcane fallacy I dig up to defend Ye. It is disappointing to witness that a magazine like Nature would allow such floppy reasoning. What is of concern, is conditional on such performance, how likely has Ye been involved in doping. The article, and the mass media, fails to give hard data to justify their accusation.
Second, on a related note, one friend on facebook pointed out: "What's interesting is that Katie Ledecky, a 15 year old, won the 800 free today. She shaved 21 second off her best time from last year. Ye Shiwen shaved 5 seconds off her best from last year (in a 400 meter race). So perhaps what Ye Shiwen did isn't that surprising." If Katie can improve 21 seconds without invoking any suspicion, it is very hard to justify accusations on Ye without resorting to racial factor (which could be justified, and will be the topic of my next blog). I am OK with taking racial factors into account, as long as 1) it is done properly (not leading to dogmatic prior) 2) it is openly acknowledged instead of denied.
Finally, the argument that test is not conclusive is red herring. The article seems to hint that now that the drug test does not rule out the possibility of Ye's doping, we should not drop our suspicion in response to the drug test result. However, every statistical test is not conclusive, and only suggestive. The key is that conditional on the result of the test, we do change our belief! At least, the likelihood of Ye doping is significantly reduced given the test result. The power of the test might not be high (which I think the article correctly point out), and we, as a result, will not as dramatically update our belief as what we would do if the power of the test is high, but to not to update our belief at all given the test result is to choose to deliberately ignore relevant evidence and information, not the trademark of a rational being. Consider this, SAT is a test for intelligence, though the power of the test is low, that is people might get higher score even if they are of mediocre intelligence. However, when you first encounter someone, who might look stupid, but you learnt that he got a really high score on SAT, you will need to update your belief about his intelligence if you are a rational being. Indeed, if you have little prior information about his intelligence (say the appearance is not a good indicator of intelligence), you will have to update your belief dramatically, despite the low power of SAT. This might sound paradoxical, but it reflects sound reasoning---in the absence of relevant information, a noisy signal, when compared to other information (that is null, and is of infinite noise), the relative noise ratio is zero! Unsurprisingly, you put all weights on this information (this is the idea behind Fisher Weighting). Hence, unless you have a pretty dogmatic prior concerning Ye, when the drug test is the only available relevant information, one would need to update his belief significantly, regardless the alleged power of the test.
These two logical flaws might be unnoticeable under first reading, but jumps out for any reader with a sense of basic (but correct) intuition about probability. It is a pity that Nature calls it fair-minded, but what is really annoying is the absence of sound logic under the pretense of sound logic.
Wednesday, August 1, 2012
More fundamental
When I go to places like Subway or convenience store in US, I am more than shocked by how inefficient the employees are. If they worked than in a private company in China, they will be laid off almost immediately or inundated by the shouts of furious customers (I admit, the Chinese customers are not polite). This is a just anecdotal evidence, but I have other reasons to believe that the US labor productivity is very low (I think partly due to the social welfare). In contrast, the labor productivity in China is much higher:
http://itechnow.com/why-apple-chose-china-to-produce-its-products/
what is described is almost unthinkable (no punch for some sports "scientist's" comments on Ye) in US. That is the difference between the US labor productivity and that of China. Not surprisingly, lots of factories decided to outsource. What is America's solution to this? For politicians, it is easy--exchange rate is the culprit. While moving exchange rate might have help balance the trade (there is no conclusive theoretical models or empirical evidence), the more fundamental solution is to improve labor productivity via enhancing education and building better incentive system (like welfare system or the power to layoff). That US never does. It likes to take superficial measures like "buy American" mandate, or appreciate RMB threat. Unless they keep "buy American" mandate forever binding, and keep depreciating USD against RMB, they will find the same imbalance coming back to haunt them. I think amid all the talks about exchange rate, this more fundamental concern on productivity is missing.
Thoughts go back to Ye again. The western response to Ye's success is just a reflection of their unhealthy psychology--passively cursing others' rise without getting their a** off and catch up. They heyday for Great Britain was long gone. America is still the top giant, but its edge has shrunk. While China still has a long long way to go (and it is not at all obvious whether it will get close to today's US), it has risen significantly. This is not taken amicably. Despite rhetoric, we dislike changes. We like things to stay where they are, what we are used to. When the balance breaks, we get uneasy. When someone who had been so frail, so weak and so despicable is catching up, there is a tinge of discomfort. Maybe it is jealousy--why am I not going up as fast? Maybe it is insecurity, will that guy threaten my place? Whatever the motive, we want to suppress that change, instead of changing with it, and improving with it. However, there will always be people who catches up, if we only think of suppressing that change, and one day, we will inevitably find we are no longer able to suppress it--then we lose our place for good, and by the way, the dignity is long lost.
http://itechnow.com/why-apple-chose-china-to-produce-its-products/
what is described is almost unthinkable (no punch for some sports "scientist's" comments on Ye) in US. That is the difference between the US labor productivity and that of China. Not surprisingly, lots of factories decided to outsource. What is America's solution to this? For politicians, it is easy--exchange rate is the culprit. While moving exchange rate might have help balance the trade (there is no conclusive theoretical models or empirical evidence), the more fundamental solution is to improve labor productivity via enhancing education and building better incentive system (like welfare system or the power to layoff). That US never does. It likes to take superficial measures like "buy American" mandate, or appreciate RMB threat. Unless they keep "buy American" mandate forever binding, and keep depreciating USD against RMB, they will find the same imbalance coming back to haunt them. I think amid all the talks about exchange rate, this more fundamental concern on productivity is missing.
Thoughts go back to Ye again. The western response to Ye's success is just a reflection of their unhealthy psychology--passively cursing others' rise without getting their a** off and catch up. They heyday for Great Britain was long gone. America is still the top giant, but its edge has shrunk. While China still has a long long way to go (and it is not at all obvious whether it will get close to today's US), it has risen significantly. This is not taken amicably. Despite rhetoric, we dislike changes. We like things to stay where they are, what we are used to. When the balance breaks, we get uneasy. When someone who had been so frail, so weak and so despicable is catching up, there is a tinge of discomfort. Maybe it is jealousy--why am I not going up as fast? Maybe it is insecurity, will that guy threaten my place? Whatever the motive, we want to suppress that change, instead of changing with it, and improving with it. However, there will always be people who catches up, if we only think of suppressing that change, and one day, we will inevitably find we are no longer able to suppress it--then we lose our place for good, and by the way, the dignity is long lost.
Thoughts on Ye Shiwen
Yeah, she won. She broke the world record. For most athletes, this would be more than fabulous, but for this 16-year old Chinese girl, this is not much short of disaster--American swimming coach called this result "disturbing" while more western media insinuate that Ye is not "clean".
Quite frankly, when I first read the news, I found the game result shocking as well. However, I was absolutely furious at people's casting suspicion that she is involved in doping. Innocent until proven guilty, easier said than done. I could not be sure if innocence holds, but before there is any evidence suggesting otherwise, I am willing and will give the benefit of doubt.
What I found more disturbing is what happened afterwards. The drug test shows no evidence of drug use while the suspicion does not there. There are talks of type II error, or somehow the Chinese developed some new drugs that evade the test (http://news.discovery.com/adventure/ye-shiwen-doping-scandal-olympic-swimming-120801.html). If one talks like this, then there is no point in employing any test--all results are non-conclusive, and whatever the test results, he can hold onto his original belief, unsupported as it might be. This is confirmation bias at play. People choose to accept evidence only if they confirm their belief; when faced with ambiguous evidence, they interpret it so that it confirms their belief. While disturbing (no punch) and obnoxious in events like this, it is dangerous in many other situations---some economists have argued that it is this fatal conceit that prepared many financial institutions for their death and damage.
The almost uniformly negative attitudes of western media towards Ye reflects a deep-rooted (and dogmatic) distrust and suspicion towards anything Chinese. It might well extend beyond sports. In politics, when things are more ambiguous, and open to more interpretations, I am wary how impartial they would be in their interpretations. As for China, I think the state is overly obsessed with its "international image". Personally, I think it is futile to worry about it and try to bolster it when the observers are (dogmatically) biased. The state's mandate is simply to improve the welfare of its people, and should not compromise this goal simply because of some "international image" concern.
Quite frankly, when I first read the news, I found the game result shocking as well. However, I was absolutely furious at people's casting suspicion that she is involved in doping. Innocent until proven guilty, easier said than done. I could not be sure if innocence holds, but before there is any evidence suggesting otherwise, I am willing and will give the benefit of doubt.
What I found more disturbing is what happened afterwards. The drug test shows no evidence of drug use while the suspicion does not there. There are talks of type II error, or somehow the Chinese developed some new drugs that evade the test (http://news.discovery.com/adventure/ye-shiwen-doping-scandal-olympic-swimming-120801.html). If one talks like this, then there is no point in employing any test--all results are non-conclusive, and whatever the test results, he can hold onto his original belief, unsupported as it might be. This is confirmation bias at play. People choose to accept evidence only if they confirm their belief; when faced with ambiguous evidence, they interpret it so that it confirms their belief. While disturbing (no punch) and obnoxious in events like this, it is dangerous in many other situations---some economists have argued that it is this fatal conceit that prepared many financial institutions for their death and damage.
The almost uniformly negative attitudes of western media towards Ye reflects a deep-rooted (and dogmatic) distrust and suspicion towards anything Chinese. It might well extend beyond sports. In politics, when things are more ambiguous, and open to more interpretations, I am wary how impartial they would be in their interpretations. As for China, I think the state is overly obsessed with its "international image". Personally, I think it is futile to worry about it and try to bolster it when the observers are (dogmatically) biased. The state's mandate is simply to improve the welfare of its people, and should not compromise this goal simply because of some "international image" concern.
Tuesday, June 26, 2012
not polished
I was reading Nancy's blog. She did not blog that much, but apparently each blog is perfectly polished. Just like her work--everytime I turn in a stack of yellow legal pads full of scrambles, she would turn in a latexed homework. My blog is a mess as well, many grammar mistakes and some spelling mistakes. I guess I take it as my notes, notes about life, not academics. I hope to jot down things and thoughts, but am too lazy to clean them up. I might lose quite a lot on the way, but cleaning them up is too time-consuming...optimization, equal MB with MC.... with this note, 70 years later, I will not look at my blog and complain I was too lazy
I have tons of notes
I have accumulated tons of notes the past year. Unfortunately, most of them are lost. Some of them actually quite valuable, and I think I put tons of effort in writing them. I still have tons left though. I printed them out and they are on my desk. But they are quite bad notes so far. So incomplete, missing details, and containing too many errors. I hope to get around and clean those notes up. Sigh...the longer I wait, the less I am able to complete those notes, and the less inclined I will be to clean those notes up. at the risk of sounding narcissist, I think they are good notes....
Subscribe to:
Posts (Atom)