Chapter 10: synthetic control has negative weights

Running `synthetic_co.w_` in the [chapter 10 notebook](https://github.com/matheusfacure/causal-inference-in-python-code/blob/main/causal-inference-in-python/10-Geo-and-Switchback-Experiments.ipynb) there are many of the weights that are coming out negative

Wondering how much to worry about this