Explanation

The White test for heteroskedasticity involves two main steps:

1. First step: Estimate the original regression model and compute the residuals
2. Second step: Regress the squared residuals on the original explanatory variables, their squares, and their cross-products, including a constant term

In this case, since there is only one explanatory variable (X₁), the correct specification for the second step should include:

• A constant term (γ₀)
• The original variable (X₁)
• The squared term of the variable (X₁²)

Therefore, option A correctly represents the second step of the White test:

$\hat{\epsilon}_i^2 = \gamma_0 + \gamma_1 X_{1i} + \gamma_2 X_{1i}^2 + \eta_i$

This specification allows testing whether the variance of the errors is related to the level of X₁ (through γ₁) and/or the squared level of X₁ (through γ₂).