Survival Analysis of United States Photovoltaic Solar Installers

The idea of this project is to study the state of household solar installers in the United States. The last couple of years say a decade has been great for them and we all have witnessed them become a normal everyday keyword from something very costly and research-oriented term. In this work, we studied more than 300000 installations made here in various location of United States and formulated it to understand the reasons for success as well as the failure of many installers in these years. In this work, we did use KM curve and Cox regression to fit a survival model. And we are also looking into network analysis feature to understand what kind of network installers formed with panel manufactures and battery manufactures over the course of ten years. The data was very dirty and I spent a lot of time cleaning it. We were able to so some basic survival analysis task but still far from our expectations. We are looking into network features, e.g., first order coupling, second-order coupling, HHI to improve upon our Cox regression model.