This is a CRR binomial tree program that prices European single-barrier knock-in calls on a dividend-paying stock, and also determines the relative error based on the call price using the Black-Scholes model.
Let c1 be the price from the tree program and c2 be the price from the Black-Scholes model. The relative error will be 100[(c1 -c2)/c2] (%)
For this project, we have:
- Inputs: S (stock price), X (strike price), H (barrier, smaller than S), t (years), s (volatility in %), r (interest rate in %), q (dividend yield in %), and n (number of periods).
- Output: Price c1 from the tree program, price c2 from the Black-Scholes model and the relative error.
In MatLab, just run the given file.
- Suppose S = 95, X = 100, H = 90, t = 1 (year), s = 25 (%), r = 15 (%), q = 5 (%), and n = 192:
- The price given by the tree is c1=5.3840.
- The price given by the Black-Scholes model is c2=5.3844.
- The relative error is -0.0083%.
- Suppose S = 95, X = 100, H = 90, t = 1 (year), s = 25 (%), r = 15 (%), q = 5 (%), and n = 193:
- The price given by the tree is c1=4.1270.
- The price given by the Black-Scholes model is c2=5.3844.
- The relative error is -23.3527%.