11. Estimation#
This chapter shifts from testing claims to estimating unknown quantities. You will use percentiles, bootstrap resampling, and confidence intervals to describe uncertainty around estimates.
Chapter Video
Video credit: CrashCourse. (2018, June 13). Confidence Intervals: Crash Course Statistics #20. YouTube. Watch on YouTube.
This embedded video is provided for educational reference only. All rights are retained by the respective copyright holder.
Learning Objectives
By the end of this chapter, you will be able to:
Identify the parameter being estimated in a business question.
Use percentiles to summarize distributions and interval endpoints.
Generate bootstrap resamples from an observed sample.
Build and interpret confidence intervals from bootstrap distributions.
Communicate estimation uncertainty without overstating precision.
Chapter Flow
Glossary
# |
Term |
Explanation |
|---|---|---|
1 |
Estimation |
Using sample data to approximate an unknown population value. |
2 |
Parameter |
A numerical feature of a population. |
3 |
Statistic |
A numerical summary computed from a sample. |
4 |
Point estimate |
A single statistic used as an estimate of a parameter. |
5 |
Percentile |
A location in ordered data; its numerical value depends on the percentile convention. |
6 |
Median |
The middle value, averaging the two middle values for even samples; conventionally the 50th percentile. |
7 |
Bootstrap |
A resampling method that samples with replacement from observed data. |
8 |
Resample |
A new sample drawn from the original sample, usually with replacement. |
9 |
With replacement |
Sampling where selected observations can appear more than once. |
10 |
Bootstrap distribution |
The distribution of a statistic across many resamples. |
11 |
Confidence interval |
An interval estimate intended to capture plausible parameter values. |
12 |
Confidence level |
The intended long-run coverage of an interval method; actual coverage depends on its assumptions. |
13 |
Margin of error |
A half-width used for a symmetric interval; bootstrap percentile intervals need not be symmetric. |
14 |
Sampling variability |
Variation in statistics across possible samples. |
15 |
Population |
The full group or process of interest. |
16 |
Sample |
The observed subset used for estimation. |
17 |
Bias |
Systematic error in an estimate or sampling process. |
18 |
Precision |
How tightly an estimate is pinned down. |
19 |
Uncertainty |
The amount of doubt around an estimate. |
20 |
Empirical distribution |
A distribution formed from observed values. |
21 |
Interval estimate |
A range of plausible values rather than a single number. |