We develop a positive behavioral portfolio theory (BPT)and explore its implications for portfolioconstrution and security design. The optimalportfolios of BPT investors resemble combinations ofbonds and lotterly tickets consistent with Friedmanand Savage's (1948) observation. We compare the BPTefficient frontier with the mean-variance efficientfrontier and show that, in general, the twofrontiers do not coincide. Optimal BPT portfoliosare also different from optimal CAPM portfolios. Inparticular, the CAPM two-fund separation does nothold in BPT. We present BPT in a single mentalaccount version (BPT-SA) and a multiple mentalaccount version (BPT-SA). BPT-SA investors integratetheir portfolios into a single mental account, whileBPT-SA investors segregate their portfolios intoseveral mental accounts. BPT-SA portfolios resemblelayered pyramids, where layers are associated withaspirations. We explore a two-layer portfolio wherethe low aspiration layer is designed to avoidpoverth while the high aspiration layer is designedfor a shot at riches.

We present large sample evidence on the performance ofdomestic and U.S. (foreign) bidder firms acquiringCanadian targets. Domestic bidders earnsignificantly positive average announcement periodabnormal returns, while U.S. bidder returns areindistinguishable from zero. Measures of pre- andpost-acquisition abnormal accounting performance arealso consistent with a superior domestic bidderperformance. Domestic bidder announcement returnsare, on average, greatest for offers involving stockpayment and for the bidders with the smallest equitysize relative to the target. Neither direct foreigninvestment controls, horizontal product marketrelationships, nor acquisition propensities explainwhy domestic bidders outperform their U.S.competitors.

We investigate the long-run return performance ofnon-U.S. firms that raise equity capital in U.S.markets. Overall, between 1982 and 1996, our sampleof 333 global equity offerings with U.S. depositaryreceipt (ADR) tranches from 35 countries in Asia,Latin America, and Europe under-perform local marketbenchmarks of comparable firms by 8%–15% over thethree years following issuance. We show thatdifferences in long-run returns are related to thescope and magnitude of investment barriers thatinduce segmentation of capital markets around theworld. While companies from markets with significantinvestment barriers for foreigners that issue equityon major U.S. exchanges outperform their benchmarks,those from segmented markets that issue equity inthe U.S. by way of Rule 144A private placementssignificantly under-perform. We also show thatinter-market competition for order flow in thepost-issuance period affects long-run returnperformance. Post-issuance buy-and-hold abnormalreturns are most significantly and positivelyrelated to the offering's ability to generate alarger share of U.S. trading volume.

This paper addresses the bias associated withparametric measurement of timing skill based onmonthly timer returns when timers can make dailytiming decisions. Simulations suggest that theclassic Henriksson-Merton parametric measure oftiming skill is weak and biased downward whenapplied to the monthly returns of a daily timer. Thepaper proposes an adjustment that mitigates thisproblem without the need to collect daily timerreturns. Four tests of timing skill, carried out ona sample of 558 mutual funds, show that very fewfunds exhibit statistically significant timingskill. More encompassing, the adjusted-FF3 test(based on the specification that incorporates boththe proposed adjustment and the Fama-Frenchthree-factor model) is the least biased measure oftiming skill among the four—it provides for asharper inference regarding timing skill and helpsmitigate biases associated with the choice ofinvestment style.

It is well known that the pro forma performance of asample of investment funds contains biases. Thesebiases are documented in Brown, Goetzmann, Ibbotson,and Ross (1992) using mutual funds as subjects. Theorganization structure of hedge funds, as privateand often offshore vehicles, makes data collection amuch more onerous task, amplifying the impact ofperformance measurement biases. Theis paper reviewsthese biases in hedge funds. We also propose usingfunds-of-hedge funds to measure aggregate hedge fundperformance, based on the idea that the investmentexperience of hedge fund investors can be used toestimate the performance of hedge funds.

Researchers are increasingly using data from the Nasdaqmarket to examine pricing behavior, market design,and other microstructure phenomena. The validity ofany study that classifies trades as buys or sellsdepends on the accuracy of the classificationmethod. Using a Nasdaq proprietary data set thatidentifies trade direction, we examine the validityof several trade classification algorithms. We findthat the quote rule, the tick rule, and the Lee andReady (1991) rule correctly classify 76.4%, 77.66%,and 81.05% of the trades, respectively. However, allclassification rules have only a very limitedsuccess in classifying trades executed inside thequotes, introducing a bias in the accuracy ofclassifying large trades, trades during high volumeperiods, and ECN trades. We also find that extantalgorithms do a mediocre job when used forcalculating effective spreads. For Nasdaq trades, wepropose a new and simple classification algorithmthat improves over extant algorithms.

This paper examines the profitability of momentumstrategies implemented on international stock marketindices. Our results indicate statiscallysignificant evidence of momentum profits. Themomentum profits arise mainly from time-seriespredictability in stock market indices—very littleprofit comes from predictability in the currencymarkets. We also find higher profits for momentumportfolios implemented on markets with higher volumein the previous period, indicating that returncontinuation is stronger following an increase intrading volume. This result confirms theinformational role of volume and its applicabilityin technical analysis.

The popular finance literature describes the assetallocation decision as one of the most importantfactors in determining investment performance. Thisarticle reviews the implications of modern portfoliotheory for the asset allocation decision and thenexaminesthe recommendations of some leadingfinancial experts to see if they are consistent withtheory.

This paper proposes a two-factor hazard rate model, inclosd form, to price risky debt. The likelihood ofdefault is captured by the firm's non-interestsensitive assets and default-free interest rates.The distinguishing features of the model arethreefold. First, the impact of capital structurechanges on credit spreads can be analyzed. Second,the model allows stochastic interest rates to impactcurrent asset values as well as their evolution.Finally, the proposed model is in closed fom,enabling us to undertake comparative staticsanalysis, compute parameter deltas of the model,calibrate empirical credit spreads, and determinehedge positions. Credit spreads generated by ourmodel are consistent with empiricalobservations.

We analyze the dividend behaviour of the aggregatestock market. We propose a model that assumesmanagers minimize the costs of adjustment associatedwith being away from their target dividend payout.The target is expressed as a function of laggedstock prices and permanent earnings, generalizingprevious models of dividend behaviour. We present anew method for measuring unobserved permanentearning based on the Kalman filter. Ourspecification of dividend behaviour is stronglysupported by the data relative to both alternativemodels and over time. We find significant evidenceof dividends conveying information regardingunexpected positive changes in current permanentearnigs. We also find that both the speed ofadjustment of dividends to target dividends andtests of signaling are sensitive to thespecification of the model.

This study directly tests the ability of severalcompeting methods to identify market buy and sellorders using intra-day quote and trade prices, andidentifies factors that affect the accuracy of themethods. Lee and Ready's (1991) algorithm performsabout the same as the tick test, but the performanceof both methods is worse than expected. The resultsshow that the use of either algorithm to classifytrades can lead to significantly biased estimates ofeffective spreads and signed volume, but the ticktest provides better estimates of effective spreadsand signed volume than Lee and Ready's method.

In this paper, I examine survivorship bias in hedgefund returns by comparing two large databases. Ifind that the survivorship bias exceeds 2% per year.Results of survivorship bias by investment stylesindicate that the biases are different acrossstyles. I reconcile the conflicting results aboutsurvivorship bias in previous studies by showingthat the two major hedge fund databases containdifferent amounts of dissolved funds. Empiricalresults show that poor performance is the mainreason for a fund's disappearance. Furthermore, Ifind that there are significant differences in fundreturs, inception date, net assets value, incentivefee, management fee, and investment styles forthe465 common funds covered by both databases.Mismatching between reported returns andthepercentage change in NAVs can partially explain thedifferences in returns.

Since hedge funds specify significant lock-up periods,we investigate persistence in the performance ofhedge funds using a multi-period framework in whichthe likelihood of observing persistence by chance islower than in the traditional two-period framework.Under the null hypothesis of no manager skill (nopersistence), the theoretical distribution ofobserving wins or losses follows a binormialdistribution. We test this hypothesis using thetraditional two-period framework and compare thefindings with the results obtained using ourmulti-period framework. We examine whetherpersistence is sensitive to the length of returnmeasurement intervals by using quarterly,half-yearly and yearly returns. We find maximumpersistence at the quarterly horizon indicating thatpresistence among hedge fund managers is short termin nature.

We distinguish between horizontal and vertical jointventures, and find correspondingly differentvaluation effects. Horizontal joint ventures createsynergistic gains that are shared by the partners.In contrast, vertical joint ventures generate gainsonly for suppliers. This is similar to the patter wefind for simple contracts, which suggests economicsimiliarities between vertical joint ventures andcontracts. Analysing firms' choices between thesecontracting options, we find that firms choosevertical joint ventures over simple contracts whenpotential hold-up problems are severe and whensuppliers face finance constraints. The results dnot support a risk-sharing motive for jointventures.

I introduce two-factor discrete time stochasticvolatility models of the short-term interest rate tocompare the relative performance of existing andalternative empiricial specificattions. I develop anonlinear asymmmetric framework that allows forcomparisons of non-nested models featuringconditional heteoskedasticity and sensitivity of thevolatility process to interest rate levels. A newclass of stochastic volatility models withasymmetric GARCH models. The existing models arerejected in favor of the newly proposed modelsbecause of the asymmetric drift of the short rate,and the presence of nonlinearity, asymmetry, GARCH,and level effects in its volatility. I test thepredictive power of nested and non-nested models incapturing the stochastic behavior of the risk-freerate. Empirical evidence on three-, six-, and12-month U.S. Treasury bills indicates but thattwo-factor stochastic volatility models are betterthan diffusion and GARCH models in forecasting thefuture level and volatility of interest ratechanges.

While theoretical models predict a decrease in the costof capital from depositary receipt offerings, theeconomic benefits of this liberalization have beendifficult to quantify, Using a sample of 126 firmsfrom 32 countries, we document a significant declineof 42% in the cost of capital. In addition, we showthe decline is driven by the ability of U.S.investors to span the foreign security prior tocross-listing. Our findings support eh hypothesisthat financial market liberalizations havesignificant economic benefits.

We examine the impact of merger announcements onprotfolios of acquiring firm and target firm commonstock from 1919 to 1930. despite vast changes in theeconomic and regulatory environment, overallacquisition profitability has remained remarkablyconstant over the last 70 and 80 years. Target firmshareholders in the 1920s clearly gained fromtakeovers, averaging abnormal retruns in excess of15%, while acquiring firm shareholders essentiallybroke even. synergistic or monopolistic gains fromconsolidation were minimal. Unlike the more recentexperience, target firm and acquiring firm abnormalreturns were largely unaffected by the mode ofacquisition, the means of financing, or the degreeof industrial relatedness.

The paper examines the determinats of contract terms onbank revolving credit agreements (revolvers) ofmedium/large publicly traded companies. We model theduration (maturity), secured status, and pricingdecisions within a simultaneous decision framework,thereby overcoming the biased and inconsistentestimates in prior single equations studies of debtcontract terms. We find strong interrelationashipsbetween contract terms with significantbi-directorinal relationships between duration andsecure status and between the all-in-spread andcommitment fees and aunidirectional relationshipfrom both duration and secured status toall-in-spread. We also illustrate how several singleequation studies of contract terms draw incorrectconclusions because of their (inappropriate)assumption that other contract therms and leveragewere exogenous. Finally, our results support thehypothesis that the setting of contract terms playsand important role in alleviating contractingproblems.

This paper argues that inferring long-horizon assetreturn predictability from the properties of vectorautoregressive (VAR) models on relatively shortspans of data is potentially unreliable. Weillustrate the problems that can arise byreexamining the findings of Bekaert andHodrick(1992), who detected evidence of in-samplepredictability in international equity and foreignexchange markets using VAR methodology for a varietyof countries from 1981–1989. The VAR predictions aresignificantly biased in most out-of-sample forecastsand are conclusively outperformed by a simplebenchmark model at horizons of up to six months.This remains true even after corrections for smallsample bias and the introduction Bayesian parameterrestrictions. A Monte Carlo analysis indicates thatthe data are unlikely to have been generated by astable VAR. This conclusion is supported by anexamination of structural break statistics. We showthat implied long-horizon statistics calculated fromthe VAR parameter estimates are very unreliable.

We investigate the value of active mutual fundmanagement by examining the stockholdings and tradesof mutual funds. We fine that stocks widely held byfunds do not outperform other stocks. However,stocks purchased by funds have significantly higherreturns than stocks they sell—this is true for largestocks as well as small stocks, and for value stocksas well as growth stocks. We find thatgrowth-oriented funds exhibit better stock selectionskills than income-oriented funds. Finally, we findonly weak evidence that funds with the best pastperformance have better stock-picking skills thanfunds with the worst past performance.