Mixed Day; Some Mathematics

Workout notes 3 mile walk (easy, West Peoria), light yoga.

Other: Today was visitation for Steve Foster, and then the burial. I was one of the pall-bearers (Steve’s wish); that I was so honored was beyond my imagination.

These things are always mixed; on one hand I liked seeing many of my athletic friends (too many to name) and I really enjoyed seeing photos of Steve that I had not seen before. Also, I was awed at how many people loved and respected him.

On the bad side, of course, is that we were all reminded at how terribly we’ll miss this guy.
Special kudos to my wife, who took off of work to be with me the entire time.

Steve’s wife Cathy did have a sense of humor though: Steve had a very rough 2007; he worked out (walking, lifting, swimming, even some yoga) but couldn’t run anymore; his last race was the Farmdale Trail Run (8 miler) on October, 14, 2006. This was his first trail race; he said that he liked it.

24 steve foster 52 m 1:17:56

There were 47 people entered at that race; I worked the finish line that day.

Steve was to find out of his illness about 1 month later; his symptoms were an acute back ache.

Anyway, Cathy noticed that Steve died just as racing season was starting, and conjectured that he could start racing again, in his life to come.

Other topics

Politics: evidently Obama means what he says when he says that he won’t accept money from registered lobbyists. A Kossak had his money returned to him for that very reason!

I just had the strangest experience. A presidential candidate gave me back my donation, told me would not accept it because of what I do for a living, and it left me more committeed to the candidate and conviced that he is the person that must be the next president.

I went to the mailbox and found a letter from the Obama Campaign. Enclosed was a check for $100, the return of my contribution from earlier this month along with a letter explaining why it would not be accepted.

You see, I am a registered lobbyist for a non-profit organziation. We are a non-partisan, non-political membership organization, we do not have a political action committee and strictly observe a policy of non-participation in any event that even remotely appears political. I serve as their legislative rep, trying to ensure that expertise of our membership is heard by public officials on issues related to their area of expertise (public safety).

I guess given the fact that I was not a corporate/industry lobbyist, I never really considered that Obama’s no-lobbyist money ban would apply to me, but it did! The letter thanked me for my interest in the campaign, but stated flately that my donation was not acceptable. [...]

Mathematics and Statistics

Role of educating engineers in statistics: an op-ed in our local paper by a Bradley engineering professor and quality control expert:

Sunday, April 6, 2008

By Dr. K.S. Krishnamoorthi

Recently, Journal Star Forum writer Dr. Sam Black wrote about the passing of quality guru Dr. Joseph Juran. Juran advanced the thesis that quality cannot be achieved with the use of statistical and other problem-solving tools alone. Management should be committed to quality. The entire organization should be focused on customer needs. Employees should be trained in quality methods.

Many U.S. businesses followed Juran’s advice and became part of the quality revolution in the late 1980s and ’90s, winning back customers they had lost to foreign competition.

The sentiment Black expresses – that the “monumental influence (of Juran) in the field of quality management will live on after him” – will be echoed by every quality professional, including this writer, who has learned from Juran’s teachings.

The mission of improving the quality of U.S.-made products is not fully accomplished. We are hearing almost the same stories that made news in the ’80s when, for example, Detroit car companies were closing plants while Japanese automakers were selling more cars. Businesses in other U.S. industries are similarly struggling with quality issues, and the fact that some of them are paying out enormous sums each year in warranty charges is but a symptom of this struggle.

It seems we need to re-evaluate the quality methodologies we employ.

In the early 1980s, when it became apparent that it was quality that helped the Japanese win markets in automobiles, electronics and other goods, many executives visited Japan to find out how their manufacturers were making quality products.

When one of those execs was asked his opinion on why U.S. manufacturers were unable to do what the Japanese were doing in relation to quality, he replied: “The average Japanese worker has a more in-depth knowledge of statistical methods than an average American engineer.”

This statement may be true even today as many quality problems, especially in manufacturing, go unresolved because there are not enough people with an in-depth understanding of statistical tools.

A recent survey by the author among quality professionals and managers showed a majority agree. They also agree that fewer than 5 percent of engineers who design and produce products have the knowledge of quality methodology and the skills to tackle the toughest quality problems.

There seems to be a general agreement among these professionals that if at least one-quarter to one-third of our engineers had a better understanding of statistics and quality tools, there would be a marked improvement in product quality.

Dr. W. Edwards Deming, another quality guru who with Dr. Juran taught the Japanese the tricks of the quality trade, said, “America needs thousands of statistically minded engineers, chemists, doctors of medicine, purchasing agents and managers.” He said training should be provided by qualified teachers who have at least a masters level education in statistics. He insisted on rigorous training in statistics theory, which would give engineers and other professionals in-depth knowledge.

As an engineering educator, I know this for a fact: Fewer than one in 20 graduates from a typical college of engineering in this country has an adequate knowledge of statistics and quality methods at graduation. Engineering educators are not even in agreement that engineers need to learn these areas.

Nonetheless, industry leaders should demand that quality education be included in the curricula of their future employees. Businesses also should train their engineers in statistical methods, with a mathematical rigor that would ensure in-depth understanding.

The quality management practices such as those recommended by Dr. Juran have greatly enhanced our ability to produce quality products. The work that remains may have to be completed with the use of statistical tools.

K.S. Krishnamoorthi is a fellow at American Society for Quality and an industrial engineering professor at Bradley University.

Disclaimer: I had professor Krishnamoorthi for two courses: quality control and a reliability engineering.

Statistics An interesting article about traditional versus Bayesian Statistics from Good Math/Bad Math.

[...]The frequentist approach to probability basically defines probability in terms of experiment. If you repeated an experiment an infinite number of times, and you’d find that out of every 1,000 trials, a given outcome occured 350 times, then a frequentist would say that the probability of that outcome was 35%. Based on that, a frequentist says that for a given event, there is a true probability associated with it: the probability that you’d get from repeated trials. The frequentist approach is thus based on studying the “real” probability of things – trying to determine how close a given measurement from a set of experiments is to the real probability. So a frequentist would define probability as the mathematics of predicting the actual likelihood of certain events occuring based on observed patterns.

The bayesian approach is based on incomplete knowledge. It says that you only associate a probability with an event because there is uncertainty about it – because you don’t know all the facts. In reality, a given event either will happen (probability=100%) or it won’t happen (probability=0%). Anything else is an approximation based on your incomplete knowledge. The Bayesian approach is therefore based on the idea of refining predictions in the face of new knowledge. A Bayesian would define probability as a mathematical system of measuring the completeness of knowledge used to make predictions. So to a Bayesian, strictly speaking, it’s incorrect to say “I predict that there’s a 30% chance of P”, but rather “Based on the current state of my knowledge, I am 30% certain that P will occur.” [...]

Follow the link to read the rest of the article and discussion.

But as far as the these to philosophies, I’ll explain it this way: the traditionalists see an event as being a truly stochastic event that has some probability of happening. The Bayesian approach is to say that “given what we know now, we can predict this event happening with probability X. But that probability might well change (that is, be different) if we knew more information.”

Think of it is way: Suppose Jones is an NBA player that has played 5 NBA basketball and has a 70% free throw shooting percentage. So when Jones goes to the line, his chance of making his first free throw is 70%. This is true whether Jones has missed his previous 10 free throws or has made 20 in a row. Why? Well, no matter what one’s free throw percentage is, one will have streaks of various kinds!

Now the Bayesian approach might be something like this: ok, Jones overall percentage is 70%, but suppose he has shot 80% over the past 10 games. Then, the probability that he makes the next free throw is probably more than 70% as there is more information (e. g., his recent free throw streak) to base your prediction on. Why might this make sense? Well, maybe he worked on his free throws, made improvements in his technique, etc.

Since I am not a statistician, I don’t have an opinion, though the Bayesian approach is considered to be the more mathematical of the two. But there are times when the traditional approach makes more sense to me: for example, it really is impossible to predict which radioactive particle will decay, no matter how much information you have. Sometimes, “god really does play dice”, so to speak.

Finally Abusing psychology: convicting someone without any direct evidence based on a psychological profile? Yes, it has happened!

DENVER (Reuters) – A man whose murder conviction was based on gruesome sketches he made as a 15-year-old boy was freed from prison on Tuesday after DNA evidence from the body was found to match the victim’s ex-boyfriend and not the convicted man.

Timothy Masters, 36, had served nine and a half years of a life sentence for the 1987 murder of Peggy Hettrick in Fort Collins, Colorado, about 45 miles north of Denver.

Masters was convicted and sentenced to life in 1999 for the crime. The mutilated body of Hettrick, 37, was found in a field near where Masters lived. He was 15 at the time.

Although police focused on the boy at the time of the killing, he wasn’t arrested until 1998, when a forensic psychologist reviewed sketches Masters had made. He was tried and convicted based on the circumstantial evidence, and an appellate court upheld his conviction and sentence.

The case was reopened last year, amid allegations that evidence was withheld from his defense team.

The case took a dramatic turn late last week, when a special prosecutor announced that traces of DNA from Hettrick’s body did not match Masters. There was, however, a match to Hettrick’s ex-boyfriend, who was dismissed as a suspect early in the case.

Master’s attorney, David Wymore, said he would ask that prosecutors dismiss the charges against his client. “I’m pretty overwhelmed and just want to see my family,” a smiling Masters told reporters after his release.

He was the first inmate in Colorado to be released from prison after DNA testing pointed to his innocence.

(Reporting by Keith Coffman in Denver; Editing by David Storey)