OpenIntro Statistics Chapter #1

OpenIntro Statistics Chapter #1

Statistics is a study of collecting and analyzing data.

Before data science, there already existed Statistics, science of collecting and analyzing data.

So, what is the difference?

Maybe where it is applied is different.

It never have been applied to business world, to be more specific, startup field, where people like to coin fancy new words.

By the way, one thing intrigued me in this chapter was a data collection part.

I never knew that regression can only be applied to data collected by random experiment.

Since my first step to data science was through machine learning, or computer science, my statistical background was bare. I applied regression to any data, whether it was collected by experiment or observations.

Therefore, in order to use regression for data collected from observations, we need to make it more alike those collected by experiment. Propensity score matching is what transforms data collected from observations to data collected from experiment, by making variables affecting the selection of treatment variable similar among treated and untreated sample.