Some Papers
     
Like Mother, Like Daughter?: A Socioeconomic Comparison of Immigrant Mothers and Their Daughters, with Margaret Hwang Smith. This paper investigates the geographic socioeconomic mobility of immigrants by comparing the residential ZIP code addresses of foreign-born mothers and their adult daughters. We find considerable movement among ZIP codes and substantial upward mobility.  
     
The Two-Child Paradox Reborn?, with Stephen Marks, Chance, 24 (1), 2011, 54-59. For at least half a century, probability devotees have puzzled over a problem known as the two-child paradox. Suppose we know that a family has two children, and we learn that one of them is a girl. What is the probability that there are two girls? Some have argued that the probability is one-half, others that it is one-third. This problem has reappeared in various guises over the years. Does it matter, for example, if we learn not only the gender of one of the children, but whether the child is the older or younger sibling? A new variant of the two-child problem argues that the two-child probabilities depend on whether the child we learn about has an unusual name, like Florida. This claim is paradoxical in that it does not seem that knowing a child’s name should have any effect on the probability that her sibling is male or female. A Bayesian approach provides a useful discipline in that it requires us to be clear about the assumptions used in our probabilistic inferences.  
     
Another Look at Baseball Player Initials and Longevity, Perceptual and Motor Skills, 112 (1), 2011, 211-216. It has been reported that Major League Baseball players whose names have positive initials (such as ACE or GOD) live an average of 13 years longer than do players with negative initials (such as ASS or BAD) or players with neutral initials (such as GHR or TSW). However, this conclusion is based on a very small sample, selective initials, and a flawed statistical test. There is no statistically significant relationship between initials and longevity for Major League Baseball players when a correct test is applied to independently selected initials.  
     
Birth Month is Not Related to Suicide among Major League Baseball Players, Perceptual and Motor Skills, 112 (1), 2011, 55-60. It has been reported that, compared to the other ten months of the year, Major League Baseball players are much less likely to commit suicide if they are born in July and much more likely to commit suicide if they are born in August. The data and statistical tests used to support this claim are both incorrect. A correct test applied to the actual data show that there is no relationship between birth month and suicide for Major League Baseball players.  
     
The Baseball Hall of Fame is Not the Kiss of Death, Death Studies, 35, 2011, 949-955. It has been reported that the median life expectancy of Major League Baseball players after election to the Baseball Hall of Fame is five years shorter than that of players of the same age who are not elected to the Hall of Fame. This conclusion is surprising because there is no compelling explanation for such a dramatic reduction in life expectancy. However, the data used in that study are incorrect. A robust test applied to correct data shows that there is no statistically persuasive difference in the life expectancy of players elected to the Hall of Fame and their peers.  
     

Poker Player Behavior After Big Wins and Big Losses, with Michael Levere and Robert Kurtzman, Management Science, 55 (9), 2009, 1547-1555. There is a growing body of evidence of psychological influences on investor decisions. A relatively unexplored question is whether experienced and generally successful investors change their investment strategy after an investment does well or does poorly. This paper investigates whether poker players change their style of play after big wins or losses. We find that experienced poker players typically change their style of play after winning or losing a big pot—most notably, playing less cautiously after a big loss, evidently hoping for lucky cards that will erase their loss. This finding is consistent with Kahneman andTversky’s break-even hypothesis and suggests that when investors incur a large loss, it might be time to take a vacation or be monitored closely. Analogous precautions might be appropriate for a manager who negotiates a bad contract, loses an important customer, initiates an unprofitable expansion, approves a disastrous advertising campaign, or endures a scandal or other public relations disaster.

  
     
Would a Stock By Any Other Ticker Smell as Sweet?, with Alex Head and Julia Wilson, Quarterly Review of Economics and Finance, 49 (2), 2009, 551-561. Some stocks have clever, eye-catching ticker symbols; for example, LUV (Southwest Airlines), MOO (United Stockyards), and GEEK (Internet America). These clever tickers might be a useful signal of the company’s creativity, a memorable marker that appeals to investors, or a warning that the company feels it must resort to gimmicks to attract investors. This paper investigates the performance of stocks with clever ticker symbols during the years 1984-2004. Surprisingly, a portfolio of clever-ticker stocks would have beaten the market by a substantial and statistically significant margin, contradicting the efficient market hypothesis.  
     

First Names and Longevity, with Laura Pinzur, Perceptual and Motor Skills, 108, 2009, 149-160. It has been reported that people’s initials can affect their life expectancy. Several studies have also reported that people with unpopular first names are perceived to be less intelligent, attractive, and likable than are people with more popular names. If so, such social stigmatization may jeopardize the life expectancy of people with unpopular names. The California Department of Health Services mortality data base for the years 1960 through 2004 for white, nonHispanic males and females was used to compare the average age at death for decedents with the most popular and least popular names.

  
     
Read Harvesting Capital Gains and Losses, with Margaret Hwang Smith, Financial Services Review, 17 (4), 2008, 309-321. Monte Carlo simulations are used to demonstrate that a very attractive tax-based trading strategy is to realize all capital losses, using excess losses to offset realized gains in order to rebalance the portfolio. This strategy increases the mean and median return by taking advantage of the tax-deductibility of losses, and mitigates risk by allowing low-cost portfolio rebalancing. This portfolio rebalancing also restarts the basis and time clock, thereby planting the seeds for a future harvesting of capital losses that can be deducted from income and used to rebalance the portfolio perpetually.  
     
Read
 “Ocular Injury Rates in College Sports,” with J. Youn, R.E. Sallis, and, K. Jones, Medicine and Science in Sports and Exercise, 40 (3), 2008, 428-432. A 2004 policy statement from the American Academy of Pediatrics and American Academy of Ophthalmology classified sports as “high risk,” “moderate risk,” “low risk,” or “eye safe” based on the number of eye injuries in each sport. However, counting the percentage of all eye injuries sustained while playing a sport creates a bias towards those sports with higher participation levels. We find that the rate of ocular injury as a result of participation in baseball, basketball, cross-country, football, golf, soccer, swimming and diving, track and field, water polo, sofiball, or volleyball is very low. Hence, any discussion with athletes regarding the utility of eye protection while participating in any of these sports should focus on the athlete's past ocular history instead of the sport to be played.  
     
“The Real Dogs of the Dow,” with Anita Arora and Lauren Capp, The Journal of Wealth Management, 10 (4), 2008, 64-72. Regression to the mean suggests that companies taken out of the Dow Jones Industrial Average may not be as bad as their current predicament indicates and the companies that replace them may not be as terrific as their current record suggests. If investors are insufficiently aware of this statistical phenomenon, stock prices may be too low for the former and too high for the latter—mistakes that will be corrected when these companies regress to the mean. Thus, stocks taken out of the Dow may outperform the stocks that replace them. We test this hypothesis with the 50 substitutions made since the Dow expanded to 30 stocks in 1928.  
     
Tobin’s q, The New Palgrave Dictionary of Economics, second edition, Larry Blume and Steven N. Durlauf, eds., Hampshire, UK: Palgrave Macmillan, 2008. Tobin’s q is the ratio of the market value of a firm to the replacement cost of its assets. This statistic can be used to predict investment spending or to control for a firm’s current and future profitability in empirical studies of corporate structure and behavior. Tobin’s q will no doubt be used in many other empirical studies of corporate structure and behavior because it circumvents the unresolved issue of how to estimate shareholders’ risk-adjusted required return by looking directly at observable market prices, which incorporate both the cash flow expectations of investors and the required returns they use to discount this anticipated cash flow.  
     
Regression to the Mean in Flight Tests, with Reid Dorsey-Palmateer, Kahnemann and Tversky report that flight trainees who do well on a maneuver typically do not do as well on the next maneuver. Because they did not understand regression to the mean, the flight instructors attributed this regression to the praise the pilots received—leading to the perverse conclusion that pilots who do well should be criticized. Kahnemann and Tversky do not report any actual data to support this memorable anecdote. This paper uses U. S. Navy flight training data to demonstrate that there is substantial regression to the mean in pilot performances. We also show how these flight scores can be used to assess changes in a pilot’s ability as the training proceeds, taking into account the anticipated regression to the mean.  
Measuring and Controlling Shortfall Risk in Retirement, with Don Gould, Journal of Investing, 16 (1) 2007, 82-95. A key challenge for retired investors is determining the stock-bond asset allocation that, for a given spending rate, provides an acceptable probability of shortfall—having real wealth drop below a specified floor during the investor’s lifetime. Standard portfolio analysis yields the well-known tradeoff between risk and return described by the Markowitz frontier. For retirement planning, we reconceptualize this as a tradeoff between shortfall probability (risk) and the median value of terminal wealth (return). For specified assumptions, there is a stock-bond asset allocation that minimizes shortfall risk. Portfolios with more stocks increase the median values of terminal wealth, but at the expense of higher shortfall risk. Portfolios with less stocks are inferior in that they decrease the median value of terminal wealth and increase shortfall risk. We find that for a variety of plausible assumptions about asset returns, investment strategies, and what constitutes a shortfall, the minimum-risk portfolio generally has between 50 and 70 percent stocks.  
     
Shrunken Interest Rate Forecasts are Better Forecasts, with Reid Dorsey-Palmateer, Applied Financial Economics, 17, 2007, 425-430. The accuracy of predicted interest rate changes can be improved by shrinking them toward a prior mean of zero. The application of this idea to interest rate forecasts by the Survey of Professional Forecasters found a consistent improvement in the accuracy of their predictions.  
     
Homeownership in an Uncertain World With Substantial Transaction Costs, with Margaret Hwang Smith, Journal of Regional Science, 47 (5), 2007, 881-896. A dynamic model of residential real estate valuation where purchases and sales are affected by uncertain projections of rents and prices.  
     
Bubble, Bubble, Where's the Housing Bubble, with Margaret Hwang Smith, presented at the Brookings Panel on Economic Activity, March 30-31, 2006; subsequently published in Brooking Papers on Economic Activity, 2006: 1, 1-50. Housing-bubble discussions generally rely on indirect barometers such as rapidly increasing prices, unrealistic expectations of future price increases, and rising ratios of housing price indexes to household income and to rent indexes. These indirect measures cannot answer the key question of whether housing prices are justified by the anticipated cash flow. We show how to estimate the fundamental value of a house and use unique rent and price data for matched single-family homes in ten metropolitan areas to illustrate this approach. These data indicate that the current housing bubble is not, in fact, a bubble in most of these cities in that, under a variety of plausible assumptions, buying a house at current market prices still appears to be an attractive long-term investment.  
     
“The Next Best Thing to Knowing Someone Who is Usually Right,” with Joseph Steinberg and Robert Wertheimer, Journal of Wealth Management, 9 (3), 2006, 51-60. Mean-variance analysis is widely used for portfolio allocation decisions. The use of historical data for the inputs may be inferior to using informed estimates that reflect one’s beliefs about the current financial environment. In this paper we show that portfolios based on expert opinion can outperform portfolios based on historical data, and that even better performance can be achieved by taking into account regression to the mean.  
     
A Great Company Can be a Great Investment, with Jeff Anderson, Financial Analysts Journal, 62 (4), 2006, 86-93. A classic investing mistake is to confuse a great company with a great investment, since a company’s well-known virtues are presumably already factored into the price of the company’s stock. We test this “mistake” by looking at the stock performance of the companies identified each year by Fortune magazine as America’s most admired companies. Surprisingly, a portfolio of these stocks outperformed the market by a substantial and statistically significant margin, contradicting the efficient market hypothesis.  
     
The Five Elements and Chinese-American Mortality, Health Psychology, 25 (1), 2006, 124-129. It has been reported that California mortality data for the years 1969-1990 indicate that Chinese-Americans are more vulnerable to those diseases that Chinese astrology and traditional Chinese medicine associate with their birth years. However, there are theoretical and statistical ambiguities in this analysis, and California mortality data for 1960-1968 and 1991-2001 do not replicate the results reported for 1969-1990.  
     
Monogrammic Determinism?, with Stilian Morrison, Psychosomatic Medicine, 67, 2005, 820-824. It has been reported that people whose names have positive initials (such as ACE or VIP) live longer than do people with negative initials (such as PIG or DIE). However, there is no compelling theory for why longevity should be markedly affected by initials and the statistical evidence of such effects is flawed by the grouping of decedents by the year of death.  
     
Regression to the Mean in Average Test Scores, with Joanna Smith, Educational Assessment, 10, 2005, 377-399. Two issues—the need to assess changes in abilities and the need to account for regression to the mean—suggest an unconventional way of using test scores to evaluate schools.  
     
Asian-American Deaths Near the Harvest Moon Festival, Psychosomatic Medicine, 66, 2004, 378-381. Independent data do not support the hypothesis that elderly Chinese-, Korean-, or Vietnamese-American women are able to prolong their lives until after the celebration of the Harvest Moon Festival.  
     
Is a House a Good Investment?, with Margaret Hwang Smith, Journal of Financial Planning, 17, 2004, 67-75. How to determine whether a house is cheap or expensive.
Bowlers’ Hot Hands, with Reid Dorsey-Palmateer, The American Statistician, 58, 2004, 38-45. Gilovich, Vallone, and Tversky’s analysis of basketball data indicates that a player’s chances of making a shot are not affected by the results of earlier shots. However, their basketball data do not control for several confounding influences. Our analysis of professional bowling indicates that the probability of rolling a strike is not independent of previous outcomes and that the number of strikes rolled varies more between games than can be explained by chance alone.
Scared to Death?, British Medical Journal, 325, 2002, 1442-1443. Are Chinese- and Japanese-Americans so frightened by the number 4 that they have abnormally high cardiac mortality on the 4th day of the month?
Idler and Kasl’s p Values: A Cautionary Lesson, Psychosomatic Medicine, 66, 2004, 373-375. A study of elderly New Haven residents indicated that some Christians and Jews postponed their deaths until after the celebration of religious holidays. However, the correct p values are larger than they report and make their conclusions less convincing, especially for Jews.

Shrunken Earnings Predictions are Better Predictions, with Margaret Hwang and Manfred Keil, Applied Financial Economics, 14, 2004, 937-943. The accuracy of analysts’ forecasts of relative earnings can be improved by shrinking their forecasts toward the mean.
The Nifty-Fifty Re-Revisited, with Jeff Fesenmaier, Journal of Investing, 11, 2002, 86–90. The basic elements of the Nifty Fifty story are sound: with the spectacular exception of Wal-Mart, the glamour stocks that were pushed to relatively high P/E ratios in the early 1970s did substantially worse than the market, in both the short and long run.
Horseshoe Pitchers’ Hot Hands, Psychonomic Bulletin & Review, 10, 2003, 753-758.. Horseshoe pitchers are more likely to throw a double ringer after having thrown a double in the preceding inning or in the two preceding innings. If their hands do not get hot, they at least get warm.
Regression to the Mean and Football Wagers, with Marcus Lee, Journal of Behavioral Decision Making, 15, 2002, 329–342.. Betting data on National Football League games indicate that gamblers do not fully account for regression to the mean.
Baseball Players Regress Toward the Mean, with Teddy Schall, The American Statistician, 54, November 2000, 231-235 (also 1999 Proceedings of the Section on Statistics in Sports, American Statistical Association, 2000, 8–13). Predictions of standardized batting averages and earned run averages can be improved consistently and substantially by using correlation coefficients estimated from earlier seasons to shrink performances toward the mean.
Baseball Careers, with Teddy Schall, Chance, fall 2000, 35–38. We investigate the typical performance record for baseball hitters and pitchers during the course of their careers. We also consider whether the length of a player’s career can be predicted accurately from his first-year performance.
Are Jewish Deathdates Affected by the Timing of Religious Holidays?, with Peter Lee, Social Biology, Spring-Summer 2000, 127–134. In contrast to other research, we find no persuasive evidence that Jews can postpone their deaths until after the celebration of religious holidays or birthdays. Our data do suggest that there may be an increase in deaths in the weeks shortly before and after birthdays.
Can the Famous Really Postpone Death?, with Heather Royer, Social Biology, Fall-Winter 1998, 302–305.. It has been reported that famous people are often able to postpone their deaths until after a birthday. A reanalysis of these data shows that there were actually a relatively large number of deaths in the month preceding and the months following the birthday.
Learning Statistics by Doing Statistics, Journal of Statistics Education, November 1998. To help students develop statistical reasoning, a traditional introductory statistics course was modified to incorporate a semester-long sequence of projects, with written and oral reports of the results. Student test scores improved dramatically and students were overwhelmingly positive in their assessment of this new approach.
Statistics for Liberal Arts Students, 1998 Proceedings of the Section on Statistical Education, American Statistical Association, 1999, 172–177. Many students will take only one statistics course in their entire lives. What should statistics professors try to teach in this one course, their only opportunity to help these students develop their statistical reasoning?
Do Statistics Test Scores Regress Toward the Mean?, Chance, Winter 1997, 42–45. Ten years of data for an introductory statistics class indicate that a student who scores one standard deviation above (or below) the mean on either the midterm or final examination is predicted to score 0.6 standard deviations above (or below) the mean on the other test.
Learning To Speak And Speaking To Learn, College Teaching, Spring 1997, 49–51. A speaking-intensive courses can not only help students develop the ability to speak coherently and persuasively, but also can help them learn a course’s subject matter.
Macroeconomic Modeling of Money, Credit, and Banking with Iman Anabtawi, Eastern Economic Journal, Summer 1994, 275-290. . A supply-and-demand model of financial markets explains several events that simpler models find paradoxical: some events stimulate the economy but contract M1; open market purchases need not be multiplied by the banking system to be powerful; business-cycle fluctuations in tax revenue can have strong effects on financial markets; and increased intermediation can be contractionary.