Option prices are assumed to contain unique information about how market participants assess the likelihood of different outcomes for future market prices. The main object of this study is to analyze the potential value of information contained in prices of options on the TAIEX index at Taiwan Stock Exchange. The information is extracted using implied risk-neutral density functions. This paper compares the quality and information contents in three alternative methods — the Black-Scholes model, the mixture of two lognormal, and the implied volatility function — by transforming these option data into the RNDs. The method based on a mixture of lognormals density is found to rank first. Despite of the different methodology in these three approaches, they lead to typical characterization of negative skewness and leptokurtosis. This implies the investors risk averse attitude in Taiwan market.