This question is regarding the Ho-Lee model:
$$ dr_t = \theta_tdt + \sigma dW_t $$
In discrete time, we can calibrate an interest rate binomial tree by finding $\theta$ in each period to match market price with the model price (expectations under the risk-neutral measure). However, every step I proceed to calibrate the next bond on the tree, the $\theta$ in the same period becomes smaller and smaller, because I'm using the calibrated interest rates for previous periods (not sure, please correct me if I'm wrong!).
My question is: what is the intuition/explanation of the $\theta$? Can we view it as the underlying "trend" of the interest rates since $\theta$ is the drift?