STAT C100 Lecture Notes - Lecture 8: Exploratory Data Analysis, John Tukey, Granularity
Document Summary
Data cleaning and manipulation: pandas, text & regexes. Kinds of visualizations and the use of size, area, and color. Data transformations using tukey mosteller bulge diagram. An introduction to database systems and sql. A model is an idealized representation of a system (cid:862)essentially, all models are wrong, (cid:271)ut so(cid:373)e are useful. (cid:863) Models enable us to make accurate predictions. A fe(cid:449) types of (cid:373)odels: (cid:862)physi(cid:272)al(cid:863) o(cid:396) (cid:862)(cid:373)e(cid:272)ha(cid:374)isti(cid:272)(cid:863) Underlying natural/causal mechanisms are assumed, though not understood. Expected to capture something about the real world (in this case) but mechanism not understood. May only capture a fleeting feature of the current state of the system. The system itself may evolve because of the use of the model. Not (cid:374)e(cid:272)essa(cid:396)ily a(cid:374)y u(cid:374)de(cid:396)lyi(cid:374)g (cid:862)t(cid:396)uth(cid:863) to (cid:271)e lea(cid:396)(cid:374)ed. Example: everyday there are some number of clouds and it rains or does(cid:374)"t. Data generation process: the real-world phenomena from which the data is collected.
