假设我在回归中使用餐厅评级作为解释变量。评级定义为， 在哪里很好，不好，是中性的并且是沉默的。我有两个概念问题。首先，我想在基于少量经验的情况下调整评级，也许通过将其缩小到整体平均值。其次，人们似乎不太愿意评价人流量大的餐厅（对于像已经存在 30 年的当地联合这样的地方来说，这是非常高的）。如果我针对, I get an inverted U shape.
Are there any transformations that I can use to remedy these two issue?
If the rating is fairly predictable based on the number of people who've been to the restaurant, it occurs to me that one possibility is to build a model of the ratings given experiences. Then you could use the residuals (obviously, constant variance is fairly crucial here) instead of the raw data.