STA130: Lecture_1

Intro to Statistics §

• Statistics - arts and science of learning from data. Its a way to quantify uncertainty
• Data Science - includes the computational aspects and acquiring, managing, and analyzing data
  • Reasoning and computing are both important aspects in data science

Statistical Thinking §

• Always be aware of the assumptions made, are they justified?
• Think about where the data came from
  • Is the data a random sample (representatives of all sample) ?
    • Survivorship Bias - only data that survived the phenomena were left, therefore these data would not be random sample
  • Are the data and the assumptions correlated?

Introduction to R and RStudio §

• Console - excites each line of code
• Source - open, view, and edit code and save for later
• Environment - list objects (variables, data frames)

• Assigning variables: using <- operator

• Comment using # symbol in R

• Element - a single piece of data, can be one fo several data types:

Data Type	R Sytax	Description	Examples
Integer	int	Numbers (integers)	-4, -2, 1, 3
Double	dbl	Numbers (floats)	2, 2.02, 22222
Logical	lgl	Booleans	TRUE, FALSE
Character	chr	Strings	“I”, “I love stats”
Factor	fct	Characters taken only out of a pre specified list	“China” out of Asian

• Vector - can be made by grouping values to the same data type (simplest data structure)

  • c() combines single elements into a vector
  • Use is. Function to check the data type [ is.numeric() , is.character()]
  • Similar to list?

• R switches between data types automatically for some operations

  • Logical -> Numeric -> Character | Logical -> Character

  > sum(c(TRUE, FALSE)) == sum(c(1, 0))
  [1] TRUE
  > sum(c(TRUE, FALSE))
  [1] 1

• Data Frame - stores data sets

  • Row: individual records; Column: variables
  • A data frame can contain multiple type of data, but within a column every cell must be the same type of data

• Packages in R provide collections of functions and data sets in addition to the things ‘base R’ can do

  • Function - a shortcut to run a bunch of code
    • Parameters - a preset setting for the expected input
    • Argument - provided input
  • tydyverse - need to be loaded into every problem set using the library() function
    • read_csv() is a function that load data in, the resulting object type is called a tibble
    • glimpse() out the number of rows and colum and listing out the column names, its data type and a few of first values (a summary out the data)
    • head() shows the top couple rows of the data (first 6 by default)

• Pipe(%>%) - a tool that makes applying functions easy (step-by-step)

  Ex. The tibble would be piped in as the input the glimpse() function

  • Tip1: read pip as “AND THEN” when reading the code
  • Tip2: use CMD + SHIFT + M to make a pipe