Difference between replicating portfolio and option price

Hello Quant Stack Exchange community, I've been working on a discrete-time model for option pricing, where I calculate the replicating portfolio using the model and compare it with the real option prices dynamically. The equation I'm using to represent this is (similar to equation of Bakshi et al. 1997 - Empirical Performance of Alternative Option Pricing Models): $$H_{t+1} = a_tS_{t+1} + C_t \cdot e^{(R \cdot \Delta t)} - P_{t+1} $$ Here, $H_{t+1}$ represents the error between the replicating p