Department of Economics
Economics 143                                              Prof. Cameron

             EXAMPLES FROM THE LITERATURE (collected Fall, 1994)

     In past years, students in Economics 143 faced with the assignment of
designing a research proposal have occasionally had some difficulty
imagining the sorts of problems for which one might use regression analysis. 
To help you recognize the vast array of issues that other researchers have
attempted to address using regression analysis, I have gathered together a
sample of citations from the recent Economics literature (downloaded from
the EconLit CD-ROM database).  Some of these studies are of very high
quality and others might reveal some problems under expert scrutiny, but all
represent efforts to use regression analysis to answer interesting

     If you are inclined to pursue the original articles corresponding to
the citations and abstracts given below, the best strategy is probably to go
to the Orion terminals at the University Research Library (URL).  Once you
have followed the posted instructions, you will type, for example,   

                   fnt  journal of banking and finance /j

This will give a list of items that have the words "journal", "banking", and
"finance" (but will limit the search to journals due to the /j option). 
Continuing to follow the Orion instructions, you will be shown in which UCLA
libraries these journals reside.  Using the STATUS command will indicate
whether the numbers you are interested in are checked out or not.  If it
looks like your item is available, proceed to the indicated library and call
number and find your copy of the bound back issue of this academic journal.


Amoako-Adu, Ben; Rashid, M.; Stebbins, M., "Capital Gains Tax and Equity
Values:  Empirical Test of Stock Price Reaction to the Introduction and
Reduction of Capital Gains Tax Exemption," Journal of Banking and Finance;
16(2), April 1992, pages 275-87.

       This paper evaluates the differential effect on stock prices of the
     introduction in Canada of &0,000 capital gains tax exemption and the
     reduction of this limit to $100,000 two years later. Using the
     seemingly unrelated regression methodology and controlling for thin
     trading and heteroscedasticity, the empirical evidence indicates that
     the changes in capital gains tax laws had a differential effect on low
     dividend yield and high dividend yield stocks on both occasions.  While
     the evidence indicates that the stock market anticipated the 1985
     capital gains tax law changes, the significant market reaction to the
     1987 reduction of the exemption occurred a day before and after the
     reading of the tax reform proposals in Parliament. Thus, it can be
     inferred from the results that, despite the presence of tax sheltering
     opportunities in Canada, changes in capital gains tax laws affect
     equity values.

Baade, Robert A.; Dye, Richard F., "The Impact of Stadiums and Professional
Sports on Metropolitan Area Development," Growth and Change; 21(2), Spring
1990, pages 1-14.

       More and more cities are being encouraged to subsidize sports
     stadiums as an economic development tool.  In this paper regression
     analysis using census data on nine different metropolitan areas is
     employed to evaluate the impact of stadiums and professional sports
     teams on area development.  Previous attempts to estimate the
     effectiveness of sports based development have used assumption driven
     trade multiplier models.  The evidence presented here is that the
     presence of a new or renovated stadium has an uncertain impact on  the
     levels of personal income and possibly a negative impact on local
     development relative to the region.  These results should serve as a
     caution to those who assume or assert a large positive stadium impact.

Baradwaj, Babu G.; Fraser, Donald R.; Furtado, Eugene P. H., "Hostile Bank
Takeover Offers:  Analysis and Implications," Journal of Banking and
Finance; 14(6), December 1990, pages 1229-42.

       A comparison of the financial characteristics of banks involved in
     hostile takeover bids with a control group of nonhostile bank mergers
     indicates that: (1) hostile targets experience abnormal returns that
     are significantly greater than for the targets of nonhostile bank
     mergers; (2) hostile bidders experience negative abnormal returns that
     are insignificantly different than for bidders involved in nonhostile
     bank mergers; (3) hostile bank acquisition announcements produce
     positive net wealth effects that are larger than the wealth effects of
     nonhostile acquisitions; and (4) a logit regression model using
     financial ratios, stock price data, and ownership data is able to
     distinguish between hostile and nonhostile targets.

Bartov, Eli, "Open Market Stock Repurchases as Signals for Earnings and Risk
Changes," Journal of Accounting and Economics; 14(3), September 1991, pages

       This paper is an empirical examination of the nature of information
     conveyed by open market stock repurchase announcements. The findings
     weakly indicate that: (1) there are positive unexpected annual earnings
     in the repurchase announcement year and positive revisions of earnings
     forecasts by analysts around announcement dates, and that (2)
     repurchase announcements are followed by declines in the repurchasing
     firms' common stock risk. In addition, a regression analysis shows that
     repurchase announcement returns are positively (negatively) correlated
     with the earnings (risk) changes conveyed by repurchase announcements.

Benjamin, John D.; Boyle, Glenn W.; Sirmans, C. F., "Retail Leasing:  The
Determinants of Shopping Center Rents," American Real Estate and Urban
Economics Association Journal; 18(3), Fall 1990, pages 302-12.

       The determinants of lease rentals are of fundamental importance to
     real estate researchers and practitioners. Retail leases are unique in
     that they typically have two rental components: a base rent and an
     "overage" rent equal to a percentage of the tenant's gross sales above
     some threshold level.  In this paper, we develop and test a simple cash
     flow model of retail lease valuation that predicts that base rents are
     lower with higher percentage rent rates and are higher with greater
     threshold levels of sales.  Using a sample of shopping center leases,
     regression analysis indicates that these trade offs are observed in the

Bergstrand, Jeffrey H., "The Heckscher-Ohlin-Samuelson Model, the Linder
Hypothesis and the Determinants of Bilateral Intra-industry Trade," Economic
Journal; 100(403), December 1990, pages 1216-29.

       Theoretical rationales for the robust empirical relationships between
     the share of intraindustry trade between two countries and the average
     levels of, and inequalities between, their GDPs, per capita GDPs, and
     tariffs have either varied or not been demonstrated formally within a
     unified analytical framework.  This study motivates theoretically the
     relationships between these six determinants, as well as the average
     level of, and inequality between, two countries' capital labor
     endowment ratios and their share of intraindustry trade.  Regression
     analysis supports, among other results, the presence of both demand and
     supply roles for per capita income influencing intraindustry trade.

Borjas, George J., "The Intergenerational Mobility of Immigrants," Journal
of Labor Economics; 11(1), Part 1 January 1993, pages 113-35.

       This article analyzes the intergenerational mobility of immigrants. 
     Using the 1940 70 censuses, the study reveals an important link between
     the earnings of immigrants and the earnings of their American born
     children. Although there is some regression toward the mean, the
     earnings of second generation Americans are strongly affected by
     variables describing economic conditions in the source countries of
     their parents. Current immigration policy, therefore, not only
     determines how immigrants perform in the labor market but also
     determines tomorrow's differences in the labor market experiences of
     American born ethnic groups.

Boyer, Marcel; Dionne, Georges; Vanasse, Charles, "Infractions au code de la
securite routiere, infractions au code criminel et gestion optimale de la
securite routiere." (With English summary.), L'Actualite Economique; 67(3),
September 1991, pages 279-305.

       In this article, the authors study the statistical relationships
     between individual probabilities of accidents and both different
     infractions to the Quebec Code of Road Safety and some infractions to
     the Criminal Code which carry a suspension of the driving permit. They
     propose new tables of demerit points based on marginal variations in
     the individual probabilities of accidents obtained from regression
     analysis and they compare their results to the Quebec official table
     (1987). They show among other results that infractions for excessive
     speed should be less penalized, that infractions associated to impaired
     (alcohol) driving and to illegal passing of school buses imply high
     probabilities of accident, and that drivers with no infractions
     represent relatively low accident risks. Their results can be used to
     improve the management of road safety. They propose three different
     methods in this article. Each can be viewed either as a substitute or a
     complement to risk based pricing of car insurance.

Browne, Lynn E., "Banks' Venture into Real Estate:  High Rollers, or
Lemmings?," New England Economic Review; 0(0), Sept. Oct. 1993, pages 13-32.

       During the 1980s, commercial banks expanded their mortgage lending
     more rapidly than other financial institutions.  This article examines
     the factors responsible for the variations in commercial banks' real
     estate lending, in an attempt to determine whether banks pursued real
     estate loans in a high risk, high return strategy, or simply were
     caught up in the general enthusiasm for real estate lending.  The
     author's regression analysis provides some support for the argument
     that banks looked to real estate loans to bolster their financial
     performance.  She also finds that in New England, where banks were
     particularly aggressive in increasing their real estate lending and
     suffered a much higher failure rate than banks nationwide, pursuit of
     real estate loans was also pursuit of growth.  And in New England,
     where most banks grew rapidly, those that grew fastest proved most
     vulnerable to failure.

Burt, Martha R., "Causes of the Growth of Homelessness during the 1980s,"
Housing Policy Debate; 2(3), 1991, pages 903-36.

       This article presents an analysis of the factors that predicted 1989
     homelessness rates in large U.S. cities. Data were collected to
     describe homelessness rates in the 182 cities with populations over
     100,000. In addition, variables were assembled to represent many
     factors that have been hypothesized to cause homelessness, including
     each city's housing and income conditions, household resources,
     employment conditions, employment structure, available public benefits,
     and cost of living. The researcher used regression analysis to assess
     the impact of each hypothesized causal factor on between city
     differences in 1989 homelessness rates for the 147 primary cities in
     the data set (excluding suburbs) and for subgroup breakouts based on
     level of manufacturing employment and population growth from 1980 to
     1986. The article ends with a discussion of policy implications of the
     patterns discovered.

Cameron, Samuel, "Race and Prosecution Expenditures," Review of Black
Political Economy; 19(1), Summer 1990, pages 79-90.

       There has been a steady accumulation of literature on racial
     differentials in crime and the response of the criminal justice system. 
     A neglected component of societal response to crime has been the volume
     of prosecution expenditures.  This article seeks to repair this neglect
     by examining state level difference in per capita prosecution
     expenditures in 1980 using a multiple regression equation.  The results
     show that, holding other factors constant, lower prosecution
     expenditures are associated with higher percentages of blacks in a
     state's population.  The relationship of this to discrimination is
     discussed in terms of England.

Cameron, Trudy Ann, "Combining Contingent Valuation and Travel Cost Data for
the Valuation of Nonmarket Goods," Land Economics; 68(3), August 1992, pages

       The travel cost method has long been used to infer the economic value
     of nonmarket resources and public goods.  More recently, contingent
     valuation survey methods have gained popularity for eliciting these
     values.  Here, contingent valuation  survey  responses are combined
     with travel cost method data on actual market behavior to estimate
     jointly both the parameters of the underlying utility function and its
     corresponding ordinary demand function.  This is a prototypical
     empirical example of a new modeling strategy, variants of which should
     prove useful in many applications, especially where reliance on a
     single valuation method is undesirable.

Cameron, Trudy Ann, "A New Paradigm for Valuing Non-market Goods Using
Referendum Data: Maximum Likelihood Estimation by Censored Logistic
Regression," Journal of Environmental Economics and Management; 15(3),
September 1988, pages 355-79.

       This paper challenges the W.M. Hanemann [Amer. J. Agr. Econom. 66,
332-341(1984)] and C. Sellar, J.P. Chavas, and J.R. Stoll [J. Environ.
Econom. Management. 13, 382-390 (1986)] utilizations of logit regression
models to estimate the value of non-market resources from 'referendum'
survey data.  These data are more informative than conventional choice data. 
The 'random utility' interpretation of logit models is therefore too
restrictive.  Bypassing the utility function entirely, it will be shown that
parameters and standard errors for utility-theoretic inverse Hicksian demand
functions can be extracted directly and much more simply.  Estimated demand
functions need not be limited to those corresponding to the linear-in-
parameters utility difference specifications which can be handled by
packaged logit regression programs.

Cameron, Trudy Ann; Englin, Jeffrey, "Cost Benefit Analysis for Non-Market
Resources: A Utility  Theoretic Empirical Model Incorporating Demand
Uncertainty," University of California at Los Angeles Department of
Economics , Working Paper: 627 , pages: 46.

       There exists a well-developed theoretical literature concerning the 
     nonmarket value of public goods under uncertainty, but little research
     effort has been devoted to utility theoretic empirical specifications. 
     This paper develops a regression-based empirical model that employs the
     state preference model of consumer decision making.  We use this model
     to assess willingness to pay for prevention of acid rain damage to
     lakes in the Northeast U.S.  Our sample includes both resource users
     and non-users and we specifically model individual participation
     decisions as a function of individual attributes, thereby allowing for
     individual risk (demand uncertainty) in the form of endogenous
     recreational participation probabilities. Controlling for user/non-user
     sample selection we use responses to a referendum contingent valuation
     survey question to calibrate an indirect utility difference function.
     We then derive the corresponding cost benefit quantities (individual
     expected consumer surplus, option price, option value, and individual
     willingness to pay loci) relevant to this application. 

Cameron, Trudy Ann; James, Michelle D., "Efficient Estimation Methods for
"Closed ended' Contingent Valuation Surveys," Review of Economics and
Statistics; 69(2), May 1987, pages 269-76.

       "Closed ended contingent valuation" surveys can be very useful in the
     evaluation of nonmarket resources.  Respondents merely state whether
     they would accept or reject a hypothetical threshold amount, either as
     payment for giving up access to the resource or as a fee for its use.
     The authors develop a maximum likelihood procedure which exploits the
     variation in the threshold values to allow direct and separate point
     estimates of regression like slope coefficients and error standard
     deviations (without truncation bias). Their illustration uses data from
     a survey of recreational fisherman to examine factors which influence
     individuals' willingness to pay.

Cameron, Trudy Ann; Quiggin, John, "Estimation Using Contingent Valuation
Data From a "Dichotomous  Choice with Follow Up" Questionnaire," Journal of
Environmental Economics and Management, (forthcoming, 1994, November)

       Dichotomous choice contingent valuation questions have gained 
     popularity over the last several years due to their purported
     advantages for avoiding many of the biases known to be inherent in
     other value elicitation formats.  However, this type of valuation 
     question is highly inefficient in that a vastly larger number of 
     observations is required to identify the distribution of values with
     any degree of accuracy. An alternative questioning strategy introduces
     a second value threshold which elicits a second discrete response. The
     size of the second offer depends on the answer to the first question.
     In analyzing data from this type of questionnaire, it is imperative to
     acknowledge the endogeneity of the second offered amount. This paper 
     demonstrates when this endogeneity is ignored or when inappropriate
     restrictions are placed upon the stochastic structure of the model. 

Cameron, Trudy Ann; Huppert, Daniel D., "OLS versus ML Estimation of Non-
market Resource Values with Payment Card Interval Data," Journal of
Environmental Economics and Management; 17(3), November 1989, pages 230-46.

       Contingent valuation methods (CVM) have been shown to be potentially
     very useful for eliciting information about demands for non-market
     goods.  We assess the sensitivity of "payment card" CVM results to the
     researcher's choice of estimation method. Empirical payment card data
     are used in both (1) a naive ordinary least squares (OLS) regression
     procedure employing interval midpoints as proxies for the true
     dependent variable, and (2) and efficient maximum likelihood (ML)
     procedure which explicitly accommodates the intervals.  Depending upon
     the design of the payment card, OLS can yield biased parameter
     estimates, misleading inferences regarding the effects of different
     variables on resource values, and biased estimates of the overall
     resource value.  (c) 1989 Academic Press, Inc.

Chew, Rosalind, "The Measurement of Gender Earnings Differentials for
Foreign Trained and Local Trained IT Professionals:  The Case of Singapore,"
Computational Economics; 6(2), May 1993, pages 95-106.

       Most studies on wage differentials have been conducted in the west.
     Few are made of the developing countries, where data on personnel
     profiles are generally found to be wanting. This study employs Borjas's
     technique to assess the effects of differences in country source
     education on gender wage differentials in a first attempt to study on
     the wage differentials of computer personnel in a non-western country. 
     Regression results indicate that the important determinants of earnings
     in the IT professions are education, total IT work experience, current
     IT work experience, age, gender, an engineering qualification, non-IT
     degrees, and overseas training. They also reveal that, male foreign
     trained professionals are paid higher salaries than male local trained
     professionals and that foreign trained female professionals in the user
     section in general also earn more than the males.

Cloninger, Dale O., "Intercity Variations in the Police Use of Lethal
Response," Journal of Economic Behavior and Organization; 17(3), May 1992,
pages 413-22.

       This paper develops and empirically tests a model of the intercity
     variation in the rate of lethal response (civilians police in
     the line of duty).  The empirically derived two stage least squares
     regression models generally confirm the relationship between crime
     seriousness and police use of force found by Robert J. Friedrich.  The
     empirical expressions in the present study explain nearly fifty percent
     of the intercity variation in the rate of lethal response and enjoy a
     high degree of statistical significance.  Specifically, the statistical
     tests indicate that the rate of lethal response is significantly and
     positively related to the seriousness of the crime, the probability of
     arrest and conviction and the age distribution  of the city's
     population.  Furthermore, there appears to be no systematic
     relationship between the rate of lethal response and the rates of
     police killed by civilians and police per violent offense.

Colton, Timothy J., "The Politics of Democratization: The Moscow Election of
1990," Soviet Economy; 6(4), Oct. Dec. 1990, pages 285-344.

       An American political scientist studying the 1989 elections in Moscow
     applies advanced analysis techniques to data gathered in the field. 
     The study is based on a uniquely large number of observations.  The
     election process,  the main political blocs fielding candidates, and
     their campaign strategies are outlined and discussed.  Regression
     analysis is used to identify relationships between election results and
     the socioeconomic characteristics of candidates as well as the
     electorate in urban neighborhoods.  The effort represents the first
     systematic attempt to trace the broad sociological contours of
     democratic elections in the U.S.S.R. on the basis of statistics
     compiled by a Western scholar.

Connolly, Catherine, "The Use of Multiple Regression Analysis in Employment
Discrimination Cases," Population Research and Policy Review; 10(2), 1991,
pages 117-35.

       This paper explores the use of multiple regression in employment
     discrimination cases brought under the Equal Pay Act of 1963 or Title
     VII of the Civil Rights Act of 1964. Through examining the outcomes of
     the cases and judicial commentary about regression analyses as
     evidence, the paper concludes that plaintiffs are most likely to
     prevail in cases where both the plaintiffs and defendants utilize
     regression. Plaintiffs who do not present regression analyses as
     evidence are at a distinct disadvantage when defendants do so. However,
     defendants are not similarly disadvantaged when responding to a
     plaintiff's regression analysis without their own statistical showing.
     Finally, plaintiffs who support their regression analyses with witness
     testimony of discriminatory employer acts appear to be more likely to
     prevail than plaintiffs who do not present such witnesses.

Culbertson, W. Patton; Bradford, David, "The Price of Beer: Some Evidence
from Interstate Comparisons," International Journal of Industrial
Organization; 9(2), 1991, pages 275-89.

       This paper is an empirical exploration into factors which are
     responsible for differences in the inter state price of beer. Though
     primarily a statistical exercise, the current work has implications for
     policy decisions.  In particular, the role of taxes and regulator
     constraints on the relationships between brewer, wholesaler, and
     retailer are highlighted.  Using an, as yet, untapped data source,
     regression analysis indicates that a  substantial proportion of the
     variance in the inter state price of beer can be explained by demand
     pressure, transportation costs, and the price of substitutes, as well
     as taxes and the presence of legally mandated exclusive territories
     governing the wholesale distribution of beer.

Davis, Lance E.; Gallman, Robert E.; Hutchins, Teresa, "Productivity in
American Whaling: The New Bedford Fleet in the  Nineteenth Century,"
National Bureau of Economic Research Working Paper: 2477 , pages: 81.

       From the end of the War of 1812 until the Civil War the New  Bedford 
     whaling fleet grew spectacularly; thereafter it declined,  equally
     spectacularly. By the end of the century New Bedford's  day was over.
     During the 88 years of this period, the technical  configuration of the
     fleet, the hunting grounds visited, and the  types of whales pursued
     all changed dramatically, and more than  once. The literature on
     whaling suggests that the collapse of  the industry was due, in part,
     to declining productivity,  occasioned by the disappearance of the
     whales (because of  over hunting) and the deterioration of the quality
     of labor. The  shifts in the composition of the fleet are viewed,
     chiefly, as  the result of efforts by whalemen to overcome their
     problems. In  this paper, productivity data (superlative indexes), by
     voyage,  are employed in multiple regression analysis to trace the 
     relationships between the changes in the composition of the  fleet and
     productivity. The propositions that declining labor  quality and whale
     stocks had important consequences for  productivity are subjected to
     test, while the impacts of  technical changes on productivity are

Dayhoff, Debra A., "High School and College Freshmen Enrollments: The Role
of Job Displacement," Quarterly Review of Economics and Business; 31(1),
Spring 1991, pages 91-103.

       The problems facing workers displaced from their jobs have received
     growing attention during the past decade.  Although economists have
     discussed the possibility that this increased risk should lead to
     greater education, no one has attempted to estimate the relationship
     between displacement threats and education enrollments.  This article
     presents empirical estimates of the determinants of high school and
     college freshmen enrollments, using the displacement rate within the
     state of residence as a regressor.  The regression results indicate
     that higher displacement rates have a positive, significant effect on
     enrollments for all groups studied.

DeBoer, Larry, "Lotto Sales Stagnation: Product Maturity or Small
Jackpots?," Growth and Change; 21(1), Winter 1990, pages 73-77.

       New York State lotto data are examined to determine whether the sales
     slowdown experienced by many state lotteries between 1985 and 1987 was
     due to a decline in interest by bettors or to a scarcity of large
     jackpots.  Regression analysis shows that lotto sales accelerate as
     jackpots grow.  Rising bettor participation causes fewer rollovers and
     smaller jackpots; this was the primary cause of the sales slowdown in
     New York.  Increasing the odds against winning generates larger prizes
     and sales growth.

Devaney, Barbara; Moffitt, Robert, "Dietary Effects of the Food Stamp
Program," American Journal of Agricultural Economics; 73(1), February 1991,
pages 202-11.

       Based on data from the 1979 80 Survey of Food Consumption in Low
     Income Households, this paper estimates the effects of changes in cash
     income and the food stamp benefit on household nutrient availability,
     while controlling for two potential sources of  selection bias.  The
     major finding of the empirical  analysis is that the estimated dietary
     effects of changes in food stamp benefits are considerably larger than
     those resulting from changes in cash income, with estimates of the
     rations of the MPC for the food stamp benefit to the cash income MPC
     ranging from three to seven across nutrients. No significant evidence
     of selection bias was found, and the estimated dietary effects of food
     stamp benefits from the selection bias models are similar to those from
     the basic model estimated by ordinary least squares regression.

Duffy-Deno, Kevin T., "Pollution Abatement Expenditures and Regional
Manufacturing Activity," Journal of Regional Science; 32(4), November 1992,
pages 419-36.

       Using a sample of sixty three SMSAs for the years 1974, 1978, and
     1982, the author finds weak support for the argument that environmental
     regulations retard economic activity. Regression analysis of the
     relationship between per unit total and air pollution abatement
     expenditures in the manufacturing sector of these SMSAs and
     manufacturing employment and earnings levels reveals that these
     associations are negative but that the magnitudes are relatively small. 
     For example, 10 percent higher per unit total pollution abatement costs
     are associated with, on average, 1064 fewer manufacturing workers (0.17
     percent of total regional employment) in Sun Belt SMSAs during 1982.

Dunlevy, James A., "On the Settlement Patterns of Recent Caribbean and Latin
Immigrants to the United States," Growth and Change; 22(1), Winter 1991,
pages 54-67.

       Between 1981 and 1987 the United States experienced a rate of
     immigration that had not been seen since before the First World War. 
     Some 25 percent of these people came from the Americas south of the
     United States.  While the great majority of these Latin and Caribbean
     immigrants tended to settle in a few select states, there are
     interesting nationality related differences in settlement preferences. 
     In this paper the settlement patterns of persons from each of eleven
     different Latin and Caribbean nations who received immigrant status in
     1987 are considered.  Regression analysis suggests that social and
     economic forces were important, but that specific factors influenced
     different nationalities differently. Evidence is also found for a
     lagged adjustment in the settlement process.  Furthermore, the
     attractive effect of a previously settled migrant stock is estimated to
     be strong for every nationality.

Easton, Peter D.; Zmijewski, Mark E., "Cross sectional Variation in the
Stock Market Response to Accounting Earnings Announcements," Journal of
Accounting and Economics; 11(2 3), July 1989, pages 117-41.

       Studies of the information content of accounting earnings typically
     assume earnings response coefficients do not vary across firms. 
     Valuation models relating earnings to security prices however, predict
     that earnings response coefficients are positively associated with
     revision coefficients (coefficients relating current earnings to future
     earnings) and negatively associated with expected rates of return.  A
     random coefficient regression model provides evidence consistent with
     these predictions.  The evidence has implications for interpreting
     multiple regression models and that relate abnormal returns to
     unexpected earnings and other information variables.

Eckard, E. Woodrow, Jr., "Advertising, Competition, and Market Share
Instability," Journal of Business; 60(4), October 1987, pages 539-52.

       This paper tests the hypothesis that advertising reduces leading firm
     market share instability (a proxy for interfirm rivalry) by creating
     market power. If advertising increases product differentiation and
     brand loyalty, then it also reduces demand cross elasticities and
     stabilizes shares. Changes over time in the combined market shares of
     the top four firms are examined for a large sample of four digit
     Standard Industrial Classification manufacturing industries for 1963
     82. Multiple regression equations are used to explain s hare
     instability, measured as deviations both about the simple mean a nd
     about a linear time trend. The results indicate that advertising does
     not reduce market share instability.

Edwards, Steven F., "Evidence of Structural Change in Preferences for
Seafood," Marine Resource Economics; 7(3), Fall 1992, pages 141-51.

       The results from graphical and regression analyses of time series
     data on seafood consumption and prices suggest that preferences for
     seafood have strengthened in response to medical evidence that seafood
     promotes nutrition and health.  The graphical analysis reveals a trend
     of increased per capita consumption of seafood since the late 1960s
     despite concurrent increases in the relative price of seafood.  The two
     phase regression analyses of per capita consumption and of the relative
     price of seafood identified the mid-1960s and the mid-1970s as possible
     times of accelerated change in preferences.  These results, which match
     those reported for consumption of poultry and red meats, have important
     implications for modeling derived demand in landings markets, for
     estimating welfare, and for managing fishing effort and multiple uses
     of fish stocks.  [This journal was not available at UCLA at last

Evans, Mark O., "An Estimate of Race and Gender Role Model Effects in
Teaching High School," Journal of Economic Education; 23(3), Summer 1992,
pages 209-17.

       This article reports an attempt to estimate race and gender role
     model effects in high school economics courses from data included in
     the National Assessment of Economic Education Survey. No evidence of a
     gender role model effect was found, but classroom role models increased
     the achievement of African American students by nearly 19 percent when
     their mothers did not have a college education. Despite the greater
     underrepresentation of black males in higher education, the black role
     model effect appears to be just as strong when the student or teacher
     is female. The study employed single equation regression models that
     included student teacher race and gender interaction variables to
     estimate the role model effects.

Figlewski, Stephen; Freund, Steven, "The Pricing of Convexity Risk and Time
Decay in Options Markets," New York University Salomon Brothers Center
Working Paper: , S 91-21 , pages: 24.

       The Black Scholes model and other related option valuation  formula 
     assume an environment in which all risk can be eliminated by a 
     continuously rebalanced hedge. In this paper we identify two  sources
     of risk in real world options that arise because hedged  positions can
     only be adjusted discretely: convexity of the  option pricing function
     and time decay. We develop measures for  the option risk associated
     with these characteristics and  examine transactions data drawn from
     the Berkeley options data  base for evidence that the market prices
     these risks. We find a  significant price effect associated with
     convexity, which  becomes larger immediately following a period of
     increased stock  price movements. The results regarding the pricing of
     time decay  were not as strong as for convexity, but were statistically 
     significant, with the hypothesized sign, in a multiple  regression that
     included the convexity measure. 

Ganderton, Philip T., "The Effect of Subsidies in Kind on the Choice of a
College," Journal of Public Economics; 48(3), August 1992, pages 269-92.

       It is possible that government subsidies to higher education,
     disbursed in kind, can produce the anomalous result of reducing the
     consumption of the good being subsidized.  In this paper, the author
     employs individual and college data to estimate the effect of such
     subsidies on students' choices of college quality.  Empirical estimates
     from a switching regression model with correction for the joint
     decisions to apply to college and choose the public or private sector
     are presented.  Predictions from the model suggest that students choose
     a substantially lower quality public college than they would have
     chosen in the private sector.

Horvath, Jane; Beaudin, Barbara Q.; Wright, Sheila P., "Persisting in the
Introductory Economics Course: An Exploration of Gender Differences,"
Journal of Economic Education; 23(2), Spring 1992, pages 101-08.

       In a logistic regression model, males are more likely than females to
     enroll in a second economics course. The grade in the first course,
     however, is shown to be a more important predictor of persistence for

Jegadeesh, Narasimhan, "Seasonality in Stock Price Mean Reversion: Evidence
from the U.S. and the U.K.," Journal of Finance; 46(4), September 1991,
pages 1427-44.

       The evidence of slowly mean reverting components in stock prices has
     been controversial.  The hypothesis of stock price mean reversion is
     tested using a regression model that yields the highest asymptotic
     power among a class of regression tests. Although the evidence that the
     equally weighted index of stocks exhibits mean reversion is significant
     in the period 1926 88, this phenomenon is entirely concentrated in
     January.  In the postwar period, both the equally weighted and the
     value weighted indices exhibit seasonal mean reversion in January.  A
     similar phenomenon is also observed for the equally weighted index of
     stocks traded on the London Stock Exchange.

Jensen, Eric R., "An Econometric Analysis of the Old Age Security Motive for
Childbearing," International Economic Review; 31(4), November 1990, pages

       A switching regression for birth interval lengths with endogenous
     switching is developed to test Mead Cain's contention that parents'
     fertility decisions arise from lexicographic preferences for old age
     security.  The model is estimated as a survival model using Malaysian
     data.  Assuming that contraceptors constitute the group of parents who
     have attained the minimal number of children with which they feel their
     old age support is secure, strong support for Cain's concept of old age
     security as the principal interest of those parents who have not yet
     attained a sufficient number of children appears in these data.

Johannesson, Magnus et al., "Willingness to Pay for Antihypertensive Therapy 
Further Results," Journal of Health Economics; 12(1), April 1993, pages

       A measurement experiment regarding willingness to pay for
     antihypertensive therapy is reported.  A new type of binary willingness
     to pay question is used that allows for different degrees of certainty
     with respect to the responses. Mean willingness to pay is derived from
     a simple expected utility model and estimated.using maximum likelihood
     methods. The estimated parameters are highly significant, with
     predicted signs, and imply a mean willingness to pay of about SEK 800
     ($130) per month. The explanatory power of the equation that only
     includes 'certain' yes/no responses is, as expected, much higher than
     that of the equation where only 'uncertain' responses are included.
     Coauthors are Per Olov Johansson, Bengt Kristrom, and Ulf G. Gerdtham.

Jones, J. C. H.; Walsh, William D., "Product Market Imperfections, Job
Content Differences and Gender Employment Discrimination at the Management
Level:  Some Evidence from the Canadian Manufacturing Sector in 1971 and
1981," Canadian Journal of Economics; 24(4), November 1991, pages 844-58.

       This paper presents a regression model of the relative employment
     level of female managers using data for 1971 and 1981 on a forty one
     industry sample drawn from the Canadian manufacturing sector.  The
     object of the analysis is to test the neoclassical proposition that
     employment discrimination, here, against females, is fostered by
     imperfect product market conditions.  In addition, the model allows for
     the open nature of the Canadian economy and for interindustry
     differences in the job and skill content of managerial occupations. 
     The results support the neoclassical view and as well emphasize the
     importance of post secondary education to the employment of females at
     the management level.

Kahn, Lawrence M., "Managerial Quality, Team Success, and Individual Player
Performance in Major League Baseball," Industrial and Labor Relations
Review; 46(3), April 1993, pages 531-47.

       This paper uses 1969 87 major league baseball data to investigate the
     impact of managerial quality on team winning and individual player
     performance. Managerial quality and player performance are measured as
     predicted pay based on salary regression s; these market based measures
     permit conclusions about costs and benefits of managerial quality.
     There are two major findings: first, when player inputs are controlled
     for, higher quality managers lead to higher winning percentages;
     second, players tend to play better, relative t o their prior
     performance levels, the higher the manager's quality.

Kang, Han Bin; Reichert, Alan K., "An Empirical Analysis of Hedonic
Regression and Grid Adjustment Techniques in Real Estate Appraisal,"
American Real Estate and Urban Economics Association Journal; 19(1), Spring
1991, pages 70-91.

       Multiple regression analysis has become increasingly popular when
     appraising residential properties for tax purposes. Alternatively, most
     fee appraisers and real estate brokers use the traditional sales
     comparison approach.  This study combines the two techniques and uses
     multiple regression to generate the adjustment coefficients used in the
     grid adjustment method.  The study compares the combined grid
     regression method with ordinary regression and defines the market
     conditions under which each method is likely to be more effective.  The
     grid regression method is found to be more accurate for relatively
     homogeneous housing markets, and the multiplicative percentage
     adjustment method (MPAM) the preferred approach.

Karafiath, Imre; Mynatt, Ross; Smith, Kenneth L., "The Brazilian Default
Announcement and the Contagion Effect Hypothesis," Journal of Banking and
Finance; 15(3), June 1991, pages 699-716.

       In this paper the authors use stock market data to test several
     hypotheses regarding possible investor reaction to the February 20,
     1987 Brazilian default. Both the traditional event study approach and a
     generalized least squares regression, using Brazilian exposure as an
     explanatory variable for abnormal returns, are used. A sample of 46
     bank holding companies, all listed on the New York of American stock
     exchanges, is divided into groups of no, low and high Brazilian
     exposure. They find that the initial stock market reaction to the
     default represents a rational investor response to new bank specific
     information. However, the stock market reaction also reveals that not
     all relevant information was contained in the disclosure announcement.
     Further, they find some evidence of a contagion effect related to size.

Kellough, J. Edward; Elliott, Euel, "Demographic and Organizational
Influences on Racial/Ethnic and Gender Integration in Federal Agencies,"
Social Science Quarterly; 73(1), March 1992, pages 1-11.

       This article examines several theoretically important demographic and
     organizational influences on the racial/ethnic and gender integration
     of federal government departments and agencies. Data on 30 government
     organizations from 1982, 1984, 1986, and 1988 are pooled to produce 120
     observations which are then analyzed with a GLS regression procedure. 
     Findings indicate that the variables analyzed explain from 54 to 70
     percent of the variation in departmental and agency work force

Linneman, Peter D.; Wachter, Michael L., "The Economics of Federal
Compensation," Industrial Relations; 29(1), Winter 1990, pages 58-76.

       The problem investigated in this paper is the economic interpretation
     of federal wage comparability laws.  We show that estimating wage
     differentials to test whether federal wages comport with the law is
     more complex than typically treated in the literature.  Although the
     issue is largely one of econometric interpretation, it addresses the
     central question of which private sector workers should be compared to
     federal workers.  We demonstrate that determining which explanatory
     variables in the wage regression equation are job descriptive and which
     are skill descriptive defines the comparison group.

Loh, Eng Seng, "The Economic Effects of Physical Appearance," Social Science
Quarterly; 74(2), June 1993, pages 420-38.

       This paper examines the wage effects of height and weight in a sample
     of full time young adult workers drawn from the National Longitudinal
     Survey Youth Cohort.  Consistent with findings in other fields on the
     effects of physical appearance, it finds that height and weight have
     statistically significant impacts on the wage level and wage growth of
     both male and female workers. Moreover, they are robust across
     regression equations.  A tentative conclusion is that these wage
     premiums reflect employer discrimination.

Lovell, Peggy A., "The Geography of Economic Development and Racial
Discrimination in Brazil," Development and Change; 24(1), January 1993,
pages 83-101.

       This study investigates the relationship between unequal resource and
     population distribution and racial wage inequalities in Brazil.  Using
     sample data from the 1980 Brazilian census, monthly wages were
     estimated for white and Afro Brazilian men working in nine metropolitan
     areas. Estimates showed that racial disparities in wages existed.across
     all regional labor markets. Regression based decomposition analysis
     found that a substantial portion of the racial wage gap was due to
     discrimination (unequal pay) while estimates of the magnitude of labor
     market discrimination indicated considerable variation by geographical
     area. Discrimination was higher in the predominantly white and highly
     developed areas of the South than in the former slave and
     underdeveloped regions of the Northeast.

Maki, Dennis; Ng, Ignace, "Effects of Trade Unions on the Earnings
Differential between Males and Females:  Canadian Evidence," Canadian
Journal of Economics; 23(2), May 1990, pages 305-11.

       This paper examines whether trade unions have affected the earnings
     gap between male and female workers in Canada.  Based on a sample of
     4,093 individuals derived from the Survey of Union Membership, the
     results of regression analysis suggest that unions have increased the
     relative earnings of male workers, even after the difference in the
     level of unionization between males and females is accounted for.  The
     results also indicate that the relative wage effect differs
     substantially between the public and private sectors.

Mankiw, N. Gregory; Romer, David; Shapiro, Matthew D., "Stock Market
Forecastability and Volatility: A Statistical Appraisal," Review of Economic
Studies; 58(3), May 1991, pages 455-77.

       This paper presents  and implements statistical tests of stock market
     forecastability and volatility that are immune from the severe
     statistical problems of earlier tests.  It finds that although the null
     hypothesis of market efficiency is rejected, the rejections are only
     marginal.  The paper also shows how volatility tests and recent
     regression tests are closely related, and demonstrates that when finite
     sample biases are taken into account, regression tests also fail to
     provide strong evidence of violations of the conventional valuation

McCrohan, Kevin F.; Pearl, Robert B., "An Application of Commercial Panel
Data for Public Policy Research:  Estimates of Tip Earnings," Journal of
Economic and Social Measurement; 17(3 4), 1991, pages 217-31.

       Unique panel data provided by NPD Research, Inc., were used to
     investigate the magnitude and interactive effects upon tipping
     aggregates and levels. The analysis revealed that tips by American
     households aggregate to $6.2, $6.3, and $6.7 billion annually in
     restaurants and other eating places in 1982, 1983, and 1984
     respectively. The information was collected by means of diaries kept by
     the tippers themselves, which was regarded as a more accurate source
     than inquiries of either the tip recipients or their employers. The
     paper covers the relationship between the characteristics of the
     tippers and of the tipping occasions and the propensity and magnitude
     of tipping. One finding of the regression analysis is that there are
     sizable differences in tip rates across tipping type restaurant
     categories when the combined effects of geographic region, metropolitan
     size, and restaurant options such as acceptance of credit cards and
     reservations and service of alcoholic beverages are considered.

McDonald, John; Shlomowitz, Ralph, "Mortality on Immigrant Voyages to
Australia in the 19th Century," Explorations in Economic History; 27(1),
January 1990, pages 84-113.

       The seaboard population of European immigrants to Australia after the
     mid-1850s was probably the first such group whose adult mortality rate
     at sea was reduced to that on land.  We calculate monthly death rates
     for various age classes and use novel regression techniques to estimate
     the correlates of mortality. For adults, the main characteristics
     associated with mortality were length of voyage and degree of crowding. 
     For children and infants, these characteristics were even more highly
     associated with mortality, as was the tonnage of the vessel.  The
     analysis highlights the special dangers faced by infants from
     infectious diseases aboard ship.  (c) 1990 Academic Press, Inc.

Mensch, Barbara; Kandel, Denise B., "Drug Use as a Risk Factor for
Premarital Teen Pregnancy and Abortion in a National Sample of Young White
Women," Demography; 29(3), August 1992, pages 409-29.

       The relationship between adolescent drug use and premarital teen
     pregnancy and abortion as a pregnancy outcome among sexually active
     women is investigated in a sample of white women from the National
     Longitudinal Survey of Youth. Event history analysis is used to explore
     whether prior drug use has a unique effect on premarital teen
     pregnancy. with controls for personality, lifestyle, and biological
     factors. Logistic regression is used to estimate whether drug use
     affects the decision to terminate a premarital teen pregnancy. The
     results show that the risk of premarital teen pregnancy is nearly four
     times as high for those who have used illicit drugs other than
     marijuana as for those with no history of any prior substance
     involvement. Furthermore, illicit drug use increases the likelihood of
     an abortion by a factor of 5.  Policy implications of the findings are

Miller, Warren B., "Personality Traits and Developmental Experiences as
Antecedents of Childbearing Motivation," Demography; 29(2), May 1992, pages

       Childbearing motivation may be conceptualized as based upon
     psychological traits and shaped by experiences during childhood,
     adolescence, and early adult life. This paper explores what those
     traits and developmental experiences are. Two measures of childbearing
     motivation, one positive and the other negative, are described. Using a
     sample of 362 married men and 354 married women, the paper
     systematically examines the factors associated with these measures. In
     addition to a set of basic personality traits, these factors include
     parental characteristics, teenage experiences, a number of variables
     from young adult behavior domains such as marriage, education, work,
     religion, and parental relationships.  Stepwise multiple regression
     analyses lead to two final constrained, simultaneous equation
     regression models. These models indicate the importance of both
     personality traits and diverse life cycle experiences in the
     development of childbearing motivation, the differential gender
     distribution of predictors, and the different experiential antecedents
     of positive and negative motivation.

Model, Karyn E., "The Effect of Marijuana Decriminalization on Hospital
Emergency Room Drug Episodes:  1975-1978," Journal of the American
Statistical Association; 88(423), September 1993, pages 737-47.

       Between 1973 and 1978 states with collectively over one third of the
     total U.S. population enacted laws that de-criminalized the possession
     of marijuana.  This article uses standard metropolitan statistical area
     (SMSA) level data on hospital emergency room drug episodes collected by
     the Drug Abuse Warning Network to measure the effect of changes in drug
     penalties on substance abuse crises.  The regression models demonstrate
     that marijuana decriminalization was accompanied by a significant
     reduction in episodes involving drugs other than marijuana and an
     increase in marijuana episodes. Although possible biases in the data
     preclude firm conclusions, the results suggest that some substitution
     occurs toward the less severely penalized drug when punishments are

Moore, Dora J., "Forecasting the Probability of Homeownership:  A Cross
Sectional Regression Analysis," Journal of Housing Research; 2(2), 1991,
pages 125-43.

       In this paper, a logit model is developed using 1985 American Housing
     Survey (AHS) data to predict the probability of homeownership for a
     given household. The variables used as regressors describe demographic
     characteristics of householders as well as family income and the
     location of the housing unit.  The probability of homeownership for
     householders with particular sets of characteristics is demonstrated. 
     Factors such as the marital status and age of the householder and the
     location of the housing unit have significant explanatory power, but
     family income remains important.  Although current demographic trends
     should have a favorable effect on the homeownership rate, economic
     policies will continue to influence household behavior.

Murayama, Yuzo, "Information and Emigrants:  Interprefectural Differences of
Japanese Emigration to the Pacific Northwest, 1880-1915," Journal of
Economic History; 51(1), March 1991, pages 125-47.

       This article examines the determinants of interprefectural patterns
     of Japanese emigration to the U.S.  Pacific Northwest, using a multiple
     regression analysis.  In estimating the regression equations, new
     proxies are introduced for the "familyfriends" effect that are free of
     the statistical problems common in previous studies on long distance
     migration. The result shows that the information networks that
     developed between pioneer immigrants and the home districts played a
     central role in shaping emigration patterns.  The lack of an
     alternative means of obtaining reliable information about conditions in
     the United States appears to be responsible.

Pastor, Manuel, Jr., "Capital Flight from Latin America," World Development;
18(1), January 1990, pages 1-18.

       This paper focuses on capital flight from Latin America during the
     1970s and 1980s.  After estimating the extent of such flight, I suggest
     that capital flight imposes growth costs, erodes the tax base, and
     worsens income distribution.  I then use regression analysis to
     investigate the causes of capital flight as well as the role of capital
     controls and IMF programs in slowing flight. I conclude by criticizing
     the current policy approach to the capital flight problems and
     suggesting some new directions.

Piette, Michael J.; Sauer, Douglas G., "Legal and Statistical Approaches to
Analyzing Allegations of Employment Discrimination," Journal of Legal
Economics; 3(1), March 1993, pages 1-14.

       This paper illustrates the legal and statistical approaches to the
     analysis of employment discrimination and details the types of issues
     generally investigated in the litigation of civil disputes where
     employment discrimination is alleged. The actual type of analysis used
     depends upon several considerations, including the size of the case,
     the particular legislation under which the case is brought, and the
     legal theory of the case.  Statistical approaches include simple Chi
     square and distributions. More complex analysis include, but are not
     limited to, OLS and LOGIT regression analysis. The expansion of
     antidiscrimination legislation into new areas, such as under the
     Americans with Disabilities Act, promises to expand the already large
     role of statistical evidence in the analysis of employment

Pilotte, Eugene, "Growth Opportunities and the Stock Price Response to New
Financing," Journal of Business; 65(3), July 1992, pages 371-94.

       The author documents the effect of growth opportunities on the stock
     price response to security offerings.  For equity offerings, the stock
     price decline for mature firms exceeds the decline for growth firms. 
     For straight and convertible debt offerings, mature firms experience a
     significant price decline while growth firms experience no significant
     price change. Regression analysis indicates that the stock price
     response to new financing is significantly, positively related to a
     variety of growth opportunity measures.  Holding growth opportunities
     fixed, the stock price response depends on the type of security offered
     (equity vs.  debt) and, for straight debt offerings, Moody's bond

Spurr, Stephen J., "Sex Discrimination in the Legal Profession: A Study of
Promotion," Industrial and Labor Relations Review; 43(4), April 1990, pages

       Although the proportion of lawyers who are women has grown rapidly in
     recent years, this study presents evidence of discrimination against
     women in promotion to partnership in major U.S. law firms in 1969 73
     and 1980.  Using regression analysis and maximum likelihood estimates
     of a probabilistic model of production of legal services, the author
     finds that women were about one half as likely as men to achieve
     partnership in those years, even though they did not significantly
     differ from men in academic distinction (selection for the law review
     or the Order of the Coif), the rank of the law schools, or

Stano, Miron; Hotelling, Harold, "Regression Analysis in Litigation:  Some
Overlooked Considerations," Journal of Legal Economics; 1(3), December 1991,
pages 68-78.

       The article shows that standard loglinear estimation of a growth
     trend, a method commonly used by experts in legal cases to forecast
     earnings and other economic magnitudes, may result in serious
     forecasting error. From earnings data introduced in a recent antitrust
     case, the results of alternative forecasting methods are compared to
     illustrate both the potential degree of error and under what
     circumstances alternatives to the standard procedure should be
     considered. The results are applicable to discrimination and wrongful
     death as well as antitrust cases.

Telford, Ted A., "Covariates of Men's Age at First Marriage:  The Historical
Demography of Chinese Lineages," Population Studies; 46(1), March 1992,
pages 19-35.

       Age at first marriage is studied in one historical Chinese population
     reconstituted from the  genealogies of 39 lineage groups for the period
     1520 1661. By using event history methods, descriptive measures of age
     at marriage for various categories of men are generated from fathers'
     ages at birth of first son as a proxy measure. The covariates of the
     likelihood of marriage at specific ages are also examined, using Cox's
     regression analysis. This study confirms an early average age at first
     marriage of 21 22 years for men, which is comparable to other
     historical Chinese populations. Some evidence for large proportions of
     celibate men and marriage in the teens for women in these lineage
     populations is also presented. Considerable variation in men's age at
     marriage is evident, primarily accounted for by differences in social

Travlos, Nickolaos G.; Waegelein, James F., "Executive Compensation, Method
of Payment and Abnormal Returns to Bidding Firms at Takeover Announcements,"
Managerial and Decision Economics; 13(6), Nov. Dec. 1992, pages 493-501.

       This study investigates the association between method of payment,
     long term performance plans, managerial stockholdings and abnormal
     returns to bidding firms at takeover announcements, using a cross
     sectional regression methodology.  Previous studies have examined each
     of these factors separately.  The results indicate that firms with long
     term performance plans and high managerial stockholdings in cash offers
     experience significantly higher abnormal returns at the announcement of
     mergers prior to 1980. The study provides additional evidence in
     explaining the previous conflicting results (Jensen and Ruback, 1985),
     examining the stock market reaction of bidding firms at merger

Truett, Dale B.; Truett, Lila J., "The Demand for Life Insurance in Mexico
and the United States:  A Comparative Study," Journal of Risk and Insurance;
57(2), June 1990, pages 321-28.

       This study compares the demand for life insurance in Mexico with that
     in the United States.  It provides a brief historical perspective on
     the growth of life insurance purchases in the two countries and employs
     regression analysis to estimate life insurance demand functions.  The
     principal findings are that age, education, and level of income affect
     the demand for life insurance and that the income elasticity of demand
     for life insurance is much higher in Mexico than in the United States.

Wagner, G. Oliver, V., "College and Professional Football Scores:  A
Multiple Regression Analysis," American Economist; 31(1), Spring 1987, pages

       No abstract available.   

Whitney, James D., "Winning Games versus Winning Championships:  The
Economics of Fan Interest and Team Performance," Economic Inquiry; 26(4),
October 1988, pages 703-24.

       Championship prospects, as distinct from game winning prospects, may
     contribute to a fan's interest in a particular sports team.  If so,
     then both season length and the structure of championship playoffs help
     determine the equilibrium allocation of playing skills across the teams
     of a league.  Evidence from a regression analysis of team attendance in
     baseball indicates that ticket demand depends, in part, on perceived
     flag winning prospects.  Several patterns in the winning percentages of
     league leaders in the major U.S. team sports are consistent with the
     perspective that championship considerations influence the allocation
     of playing skills.

Williams, Mary L.; Waldauer, Charles; Duggal, Vijaya G., "Gender Differences
in Economic Knowledge:  An Extension of the Analysis," Journal of Economic
Education; 23(3), Summer 1992, pages 219-31.

       This article extends the analysis of gender differences in economic
     knowledge by examining college students in intermediate economic theory
     and  economic statistics courses, as well as in principles of
     economics. In addition, this investigation includes mathematical
     problems, graphs, and essay questions, along with the standard multiple
     choice questions in the tests administered. Using ordinary least
     squares multiple regression analysis, no consistent evidence is found
     to support the hypothesis that significant gender differences exist in
     college student performances on economic exams.  Also, there is no
     indication that males perform better on questions requiring
     quantitative skills, nor that females perform better on questions
     requiring verbal skills.

Yang, Bijou, "The Economy and Suicide:  A Time Series Study of the U.S.A.,"
American Journal of Economics and Sociology; 51(1), January 1992, pages

       The suicide rate for the United States for the period of 1940 84 was
     posited to be the consequence of the interplay of economic and social
     variables. The single equation  regression was applied to the suicide
     rates for the total population and for the four sex by race social
     groups. The results indicated that: (1) suicide rates did not increase
     during the economic booms and busts as predicted by Durkheim, and the
     change depended upon the social groups involved; (2) the unemployment
     rate had significant detrimental impact only on the white male suicide
     rate; (3) the female labor force participation rate had beneficial
     impact on both the white and non-white female suicide rates; (4) the
     divorce rate was the only variable that had a consistent impact for all
     social groups; and (5) membership in the Catholic Church had a positive
     association with the suicide rates.