If time and computing facilities are available, consider the following 40 losses, due to wind-related catastrophes, that were recorded to the nearest $1 million (these data include only those losses of $2 million or more, and, for convenience, they have been ordered and recorded in millions of dollars): 2222222222 2233334445 5556666889 15 17 22 23 24 24 25 27 32 43 To illustrate bootstrapping, take resamples of size n = 40 as many as N = 100 times, computing the value of T = ( X 5)/(S/ 40 ) each time. Here the value 5 is the median of the original sample. Construct a histogram of the bootstrapped values of T.
Section 3: Statistics in Government Stat 2220, Espey ➢ The US has 72 federal statistical offices that find statistics such as unemployment rate, health statistics, statistics of judicial cases, the census, and income ➢ Qualitative Variable Statistics ○ Sample Proportion: A value between 0 and 1 or a percentage ︿ Total number of successex ■ p = Total number of observatinns ○ Ex: the unemployment rates show how many people out of the entire labor force are unemployed. People do not count as unemployed if they have not looked for a job in the last four weeks. They also do not count as unemployed even if they are only working 1hr/week or getting paid $20/week or are underemployed. ■ Establishment Survey: measures the number of jobs by gathering data from 400,000 companies and estimates the number of Americans with payroll jobs from that ■ Household Survey: data is gathered from 60,000 households from which the unemployment is estimated (includes those who are self employed or paid under the table) ○ Ex: Labor force participation rate=mployed + Unemployed population ➢ Quantitative Variable Statistics ○ Ex: Income ○ Measures of Central Tendency: interested in the most typical observation (center) Σx1 The sum of all observations ■
