The F-statistic for testing the significance of the joint hypothesis \( H_0: \beta_3 = \beta_4 = \beta_5 = 0 \) is given by:
\[
F = \frac{(R^2_1 - R^2_2)/p}{(1 - R^2_1)/(n - k_1)}
\]
where:
- \( R^2_1 = 0.3521 \) is the \( R^2 \) value from model (1),
- \( R^2_2 = 0.2314 \) is the \( R^2 \) value from model (2),
- \( p = 3 \) is the number of restrictions (since \( \beta_3, \beta_4, \beta_5 \) are being tested),
- \( n = 350 \) is the sample size,
- \( k_1 = 5 \) is the number of predictors in model (1).
Substituting these values into the formula:
\[
F = \frac{(0.3521 - 0.2314)/3}{(1 - 0.3521)/(350 - 5)} = \frac{0.1207/3}{0.6479/345}
\]
\[
F = \frac{0.0402}{0.00188} \approx 21.34
\]
Thus, the value of the test statistic is \( 21.340 \).