A More Formal Treatment of Classification and Forecasting
This chapter covers much of the foundational material from the last chapter but more formally and in more detail. The chapter opens with a discussion of the model used to characterize how the data were generated. That model is very different from the one used in conventional regression. Then, classification is considered followed by the estimation issues it raises. The bias-variance tradeoff is front and center. So is post-selection statistical inference. Some material in this chapter may require careful reading.
- Berk, R. A., Bjua, A., Brown, L., George, E., Kuchibhotla, A.K., Sue, W., & Zhau, L. (2018b) Assumption lean regression. arXiv:1806.09014v1 [stat,ME].Google Scholar
- Buja, A., Berk, R., Brown, L., George, E., Pitkin, E., Traskin, M., Zhan, K., & Zhao, L. (2018a). Models as approximations — part I: a conspiracy of nonlinearity and random regressors in linear regression.” arXiv:1404.1578Google Scholar
- Kuchibhotla, A., Brown, L., Buja, A., George, E., & Zhao, L. (2018) A model free perspective for linear regression: uniform-in-model bounds for post-selection inference. arXiv:1802.05801v2 [math.ST]Google Scholar
- Rice, J. A. (2006) Mathematical Statistics and Data Analysis. Third Edition. New York: Duxbury Press,Google Scholar