Academic journal article International Journal of Business

Measuring the Systematic Risk of IPO's Using Empirical Bayes Estimates in the Thinly Traded Istanbul Stock Exchange

Academic journal article International Journal of Business

Measuring the Systematic Risk of IPO's Using Empirical Bayes Estimates in the Thinly Traded Istanbul Stock Exchange

Article excerpt


The systematic risk of IPO's in the thinly traded Istanbul Stock Exchange (ISE) are estimated using Empirical Bayes Estimators (EBE). The sectors that the firms belong to, provide the priors. Comparisons are made with OLS estimators across different estimation and forecasting periods. Two benchmark criteria are used; sum of squared residuals and sum of absolute residuals. The application requires some complicated manipulation of the theory where some inferiors of the ordinary Bayesian approach are avoided. Results show that using the EBE procedure, betas can be calculated with greater precision than OLS. This enables us to evaluate IPO's on similar intuition with other stocks, i.e. in a portfolio context rather than in isolation.

JEL: C2, C11, C52, G1, G12

Keywords: Empirical Bayes method; Beta estimation; Forecasting; Capital Asset Pricing Model; Initial public offering.


The capital-asset-pricing model (CAPM) of Sharpe (1964) and Litner (1965) constitutes a cornerstone in finance literature. CAPM is still popular among academics as well as practitioners in estimating expected returns. It has a powerful intuition that centers on the assumption that investors hold mean-variance efficient portfolios as described by Markowitz (1959) and the underlying economics is clear-cut. Estimation of expected returns using CAPM is simple and straightforward.

Almost all textbooks in corporate finance recommend CAPM expected returns for estimating the cost of capital. Academics in empirical finance typically use CAPM to estimate benchmark expected returns. Despite the sound theoretical foundations and, simple estimation procedure, empirical problems remain. Considerable empirical evidence is reported that the market [beta] alone is not sufficient to describe the expected returns on individual securities (1). Several multi-factor alternatives have been offered the most prominent of which is the Arbitrage Pricing Theory of Roll and Ross (1984). Also, it is known that tests of CAPM are sensitive to the proxies used for market portfolio (Stambaugh, 1982) and the relationship between stock returns and systematic risk contains non-linearities (Tinic and West, 1986). Still, the market model is widely used as a benchmark for performance evaluation and for measurement of abnormal returns in event studies, due to its strong intuition and straightforward estimation.

One way to circumvent the trade-off between empirical problems in describing expected returns and the practical need for a straightforward measure of systematic risk is to use alternative estimation procedures. Depending upon the needs of the researchers, different estimation procedures have been proposed. Fama-MacBeth (1973) two-pass regressions and Fama-French (1992) three factor model are the most well known. Karolyi (1992) argues that adjustment techniques based on cross-sectional information are uninformative and introduces prior information in the form of size and sector based cross-sectional distributions. Berra and Kannan (1986), suggest a multiple root-linear model to adjust the betas, assuming that betas are changing over time with a regular pattern towards the mean value. Vazicek (1973) proposed a Bayesian adjustment technique where the weighted-average of historical and cross-sectional betas is used. Dimson (1979) considered the case of infrequent trading and proposed an aggregated coefficients method for estimating the betas. Siegel (1995) estimates betas from observed option prices to account for the implicit volatility of the stock. Wittkemper and Steiner (1996) use neural networks to predict the betas. This way, they account for the non-linear interdependencies between a number of variables besides the past returns. In that sense, they actually employ a multi-factor model.

The purpose of this paper is to introduce an alternative estimation procedure, Empirical Bayes (EB) Estimates, to tackle another problematic case, the initial public offerings, (IPOs). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.