Canonical Least-Squares Monte Carlo : Empirical Evidences from S&P 100 Index and IBM Puts

This paper tests empirically the recently proposed canonical least-squares Monte-Carlo (CLM) method for pricing American options. Market data used are daily last prices for the American style S&P 100 Index puts (OEX puts) and IBM puts from 30 July 2008 to 30 January 2009, the period of which covers the US financial crisis of 2008. Unlike what have been done in many other similar empirical studies, rates interpolated from the US Treasury yield curve are used, and dividends, a continuous constant yield for the S&P 100 Index and a fixed cash amount for the IBM common stock, are taken into consideration. Compared with the benchmark prices from finite difference using historic volatilities, CLM outperforms in almost all twelve categories of moneyness and maturity for both OEX and IBM puts, and is shown to be a viable alternative for pricing American style options