A researcher is interested in studying relationship between credit rating and ownership structure of public companies in Thailand by using the following linear regression model: Rating: = Bo + Bifami + Bz In(debt,) + Byfam;. In(debt,) + U; where Rating is credit-rating score ranging from 0 to 100 points with higher points mean better creditworthiness, fam is the dummy variable taking value 1 if it is a family-owned business, and debt is outstanding debt in unit of million Baht. Suppose that the researcher wants to interpret this model as ceteris paribus causation, and he can justify that all the regressors are uncorrelated with the error term. He uses an lid data set and STATA to get OLS estimates as follows gen 1debt = 1n(debt) gen x2 - fam*1debt gen x3 - x2 - 1debt reg rating fam x2 x3 roa Coef. Std. Err. fam -4.200 2. 100 x2 -70.920 35.460 69.670 18.000 _cons 68.000 23.000 test x2 + x3 = 0 ( 1) x2 + x3 = 0 F(1, 600) - 6.25 prob > F = 0.0126 i. (8 points) Conduct a hypothesis test at 10% significance level to test whether an increase in outstanding debt has negative impact on credit rating for family-owned businesses. State null and alternative hypotheses, test statistics, distribution of the test statistics including the degree of freedom, p-value, critical value and the conclusion of the test