请问R SQUARE 和异方差的关系是什么？

问题如下：

In a model with a single explanatory variable, what value of the $R^2$ in the second step in a test for heteroskedasticity indicates that the null would be rejected for sample sizes of 100, 500, or 2500?

(Hint: Look up the critical values of a $\chi_q^2$, where q is the number of restrictions in the test.)

选项：

解释：

When there is a single explanatory variable in the original model, the auxiliary regression used for White’s test will have an intercept and two explanatory variables—the original variable and the variable squared. White’s test statistic is $nR^2$, and the null has a $\chi_2^2$ distribution. When using a test with a 5% size, the critical value is 5.99. Solving for the $R^2$, $R^2 \leq5.99/n$. The maximum value of $R^2$ that would not reject is 0.0599, 0.011, and 0.0023 when n is 100, 500, and 2500, respectively.