Читать книгу Business Experiments with R - B. D. McCullough - Страница 4

List of Illustrations

Оглавление

1 Chapter 1Figure 1.1 Life expectancy vs. newspapers per 1000 (left) and log(newspapers...Figure 1.2 Boxplot of default vs. non‐default for credit limit and ages.Figure 1.3 Sample selection bias. Dashed line for observations and solid l...Figure 1.4 A/B test for mailing list sign‐ups.Figure 1.5 Skirt images test.Figure 1.6 Mobile landing page test for storage company.Figure 1.7 Video icon test.Figure 1.8 US agricultural output and input in the twentieth century.Figure 1.9 US car imports from Japan.

2 Chapter 2Figure 2.1 Box plot of one‐week sales resulting from audio versus video sale...Figure 2.2 Unstacked (left) and stacked (right) data.Figure 2.3 Mosaic plot of email response data.Figure 2.4 Box plots comparing sales for alternative landing pages. Left pan...Figure 2.5 Beta distributions.Figure 2.6 Beta‐binomial distribution (left) and blowup of region (right)....Figure 2.7 Example report from Google Analytics Experiments shows analysis b...Figure 2.8 Display ads for Acme hotel, versions A (left) and B (right).Figure 2.9 Wald interval coverage for a nominal CI.

3 Chapter 3Figure 3.1 Range of and precision of slope estimates.

4 Chapter 4Figure 4.1 One‐sided, two‐sample test.Figure 4.2 One‐sided power calculation.Figure 4.3 Box plot and dot plot of typing test performance.Figure 4.4 Before and after typing scores.Figure 4.5 Box plot for the car rental data.Figure 4.6 Tukey HSD intervals for the car rental data.Figure 4.7 MDES when and : .Figure 4.8 Subgroup box plots for landing pages (zeros excluded from data)....Figure 4.9 An illustration of Simpson's paradox in regression.Figure 4.10 Null distribution for a one‐sided, one‐sample test.Figure 4.11 A two‐sided interval is the intersection of two one‐sided interv...Figure 4.12 Lower bound interval where .

5 Chapter 5Figure 5.1 Boxplots of call center data.Figure 5.2 Plot of call center data.Figure 5.3 DigiPuppet on a child's finger.Figure 5.4 Box plots for separate distances.Figure 5.5 Box plots for combined distances.Figure 5.6 Randomization vs. rerandomization.Figure 5.7 Harry's online banner (left) and mobile Facebook (right) ads.Figure 5.8 Slushie sales by week. Control, solid line; treatment, dashed lin...

6 Chapter 6Figure 6.1 Square diagram for simple experiment.Figure 6.2 Interaction plot for interest rate and annual fee.Figure 6.3 Interaction plot for data in Table 6.9.Figure 6.4 Square diagram for main and interaction effects for data in Table...Figure 6.5 Main effect plots for loan experiment.Figure 6.6 Two interaction plots for loan experiment.Figure 6.7 Two‐way interaction plots for credit card example.Figure 6.8 Half‐normal plot for the fully saturated credit card experiment....Figure 6.9 Score vs. MSAT.Figure 6.10 Pairs of orthogonal vectors.

7 Chapter 7Figure 7.1 Half‐normal plot of postcard data.Figure 7.2 Two parallelepipeds.Figure 7.3 All possible design points.Figure 7.4 Determinants and areas of two three‐point designs.Figure 7.5 A boring color map of the correlations.Figure 7.6 Average pizza sales by day of week.Figure 7.7 Color maps for Bayesian D‐optimal design.Figure 7.8 Color maps for minimal aliasing design.Figure 7.9 Determinants and areas of various 3‐point designs.

8 Chapter 8Figure 8.1 Color map of the correlations for fractional factorial.Figure 8.2 Color map of the correlations for six factors in 16 runs, fractio...Figure 8.3 Color map of the correlations for six factors in 12 runs, Placket...Figure 8.4 Color map of ideal screening design.Figure 8.5 Color map of a really good screening design.

9 Chapter 9Figure 9.1 Output of optFedorov command.Figure 9.2 Color map for Taurus design.Figure 9.3 Color map for 14 runs.Figure 9.4 Color map for blocked Taurus design.Figure 9.5 Color maps for four factors in eight runs (left is from AlgDesign Figure 9.6 Irregular design region.

Business Experiments with R

Подняться наверх