A regression model is estimated using the equation $Y = b_0 + b_1X_1 + b_2X_2 + b_3X_3 + \varepsilon$, where regression statistics show that $X_1$ and $X_2$ are highly statistically significant but $X_3$ is not statistically significant. Which of the following statements about the use of this model for prediction is most accurate?