Kian Guan Lim

A New Test of the Three-Moment Capital Asset Pricing Model

This paper tests the Kraus-Litzenberger (1976) three-moment capital asset pricing model using Hansens (1982) generalized method-of-moments (GMM). The GMM approach does not impose strong distributional assumptions on the asset returns. This is an interesting issue since there is no obvious multivariate distribution for returns that also exhibits co-skewness. Using monthly stock returns to test the model, there is some evidence that systematic skewness is priced.

Portfolio value-at-risk optimization for asymmetrically distributed asset returns

We propose a new approach to portfolio optimization by separating asset return distributions into positive and negative half-spaces. The approach minimizes a newly-defined Partitioned Value-at-Risk (PVaR) risk measure by using half-space statistical information. Using simulated data, the PVaR approach always generates better risk-return tradeoffs in the optimal portfolios when compared to traditional Markowitz mean–variance approach. When using real financial data, our approach also outperforms the Markowitz approach in the risk-return tradeoff. Given that the PVaR measure is also a robust risk measure, our new approach can be very useful for optimal portfolio allocations when asset return distributions are asymmetrical.

Information-time option pricing: theory and empirical evidence

Abstract With a stochastic time change from calendar-time to information-time, we derive a parsimonious option pricing formula with stochastic volatility as a risk-neutral Poisson sum of Mertons (1973) prices over the options information-time maturity domain. The formula contains two unobservable parameters, information arrival intensity and information-time asset volatility, with stochastic volatility induced by random information arrival. When the information arrival rate intensifies, the option price increases and vice-versa. We test the formula in pricing, hedging, and excess profits capture empirically using currency and the S&P 500 futures options transaction data.

Computing maximum smoothness forward rate curves

Adams and Van Deventer (1994) presents a new approach1 to yield curve smoothing that provides a sound basis for implementing many of the no-arbitrage term structure models such as Vasicek (1977), Heath, Jarrow and Morton (1992), Hull and White (1993) and so on. By carefully defining the criterion for the best fitting yield curve to have maximum smoothness for the forward rate curve, they arrive at a simple but powerful method providing a closed-form solution for a yield curve that fits all observed yields. A key result in that paper that enables this method or procedure is that the smoothest forward rate curves are produced by a fourth-degree polynomial with the cubic term missing. In this paper we show that excluding the cubic term is incorrect and sup-optimal. We provide a correct proof of the result that the smoothest forward rate curves are produced by an unconstrained fourth-degree polynomial. We also provide the numerical algorithm to compute the corresponding yield and the forward rate curves.

Tests of Rational Bubbles Using Cointegration Theory

This paper provides results of tests of cointegration of stock price and stock dividend payments. Taking into account the problems of nuisance parameters and size distortion in the test statistics used in these tests, we find evidence that stock price and stock dividends are not cointegrated. More specifically, we have not ruled out the presence of rational bubbles.

Information Transmission across Eurodollar Futures Markets

Identical contracts traded at two distinct time zones but linked with a mutual offset system allow for round-the-clock trading. It is interesting to investigate the characteristics of information transmission in such a market. In this paper we employ GARCH specifications to model Eurodollar futures interest rate changes in IMM and in SIMEX. We employ the Causality-in-Variance test based on cross-correlation function to test hypotheses on the lead-lag relationships of volatilities between IMM and SIMEX. There is strong evidence of information transmission from IMM to SIMEX, and not vice-versa. However, during the 86 December till 88 September period, including contracts during the October 87 crash, there is evidence of causality also running from SIMEX to IMM.

Portfolio hedging and basis risks

Minimum variance hedged portfolios using futures are formed by taking the linear projection of spot price changes onto futures price movements as the hedge ratio. This unwittingly assumes that the underlying spot-futures price movements follow a cointegrated process, given that the spot and the futures prices are integrated processes. If the spot-futures prices are not cointegrated, the hedged portfolio suffers from the risk of potentially large changes in its value. Empirical findings using the Nikkei stock index and the Nikkei 225 futures show this deviation in intraday trading prices. The basis movements which have often been used by intraday traders to predict future price changes, are tested to be mostly unit root processes. This is shown to be due largely to non-cointegration of the spot-futures prices, and suggests why it is profitable to trade futures using basis knowledge only if trading is done on a continual basis.

Financial valuation and econometrics

This book brings together domains in financial asset pricing and valuation, financial investment theory, econometrics modeling, and the empirical analyses of financial data by applying appropriate econometric techniques. These domains are highly intertwined and should be properly understood in order to correctly and effectively harness the power of data and methods for investment and financial decision-making. The book is targeted at advanced finance undergraduates and beginner professionals performing financial forecasts or empirical modeling who will find it refreshing to see how forecasting is not simply running a least squares regression line across data points, and that there are many minefields and pitfalls to avoid, such as spurious results and incorrect interpretations.

An approximation pricing algorithm in an incomplete market: A differential geometric approach

Abstract.The minimal distance equivalent martingale measure (EMM) defined in Goll and Rüschendorf (2001) is the arbitrage-free equilibrium pricing measure. This paper provides an algorithm to approximate its density and the fair price of any contingent claim in an incomplete market. We first approximate the infinite dimensional space of all EMMs by a finite dimensional manifold of EMMs. A Riemannian geometric structure is shown on the manifold. An optimization algorithm on the Riemannian manifold becomes the approximation pricing algorithm. The financial interpretation of the geometry is also given in terms of pricing model risk.

Information and liquidity effects of government approved stock investments

Abstract On April 1, 1986, the Singapore government announced that resident investors or Central Provident Fund (CPF) members could use their compulsory savings in the CPF to invest in approved equity stocks, unit trusts, and gold. This policy was implemented on May 1, 1986. However, the list of 70 approved trustee stocks in which CPF members may invest their funds was not released until May 3, 1986. This paper studies the economic impact of the event of releasing the list of 70 approved trustee stocks on the returns of both the approved and the non-approved stocks. Two possible impact are suggested, namely the information effect and the liquidity constraint effect. A multivariate regression model as well as a traditional event study time series model are used to test for the presence of the two effects. We test for both the partial information release on April 1, and also the full information release on May 3. The results show that significant information effect was not present. However, there is evidence that liquidity constraint did exist before May 1, 1986. The paper draws two implications which may be useful to policymakers. Firstly, the releasing of information in stages can effectively reduce informational impact on the stock market. Secondly, the existence of a liquidity constraint implies that compulsory saving schemes may cause investors, the savers, to hold more riskfree assets than they really desire.