2. Derive the composite midpoint method and composite error.