7. With inear models, we make relatively strong assumptions about the underlying rela- tionships and distributions and these assumptions allow us to provide standard inferential results (confidence intervals and hypothesis tests). Now w look at how we could carry out corresponding versions of these with permutation tests. That is, we'll think about how we could try and ask the same types of scientific questions (i.e. evaluate the same kinds of statistical hypotheses) without relying on these known results (a) For now, we'l look at multiple linear regression with just two predictors. Let's start by generating some data. Generate a training dataset with 50 observations using the following code x1