Index

Note: Page numbers followed by b indicate boxes, f indicate figures and t indicate tables.

A Bayesian hypothesis testing

Absolute error loss function, 559 Jeffreys’ hypothesis testing criterion, 574
a-trimmed mean, 305 null hypothesis, 573
Alternate hypothesis, 313 posterior odds ratio, 574
Analysis of variance (ANOVA) posterior probability, 573
angular transformation, 545 prior odds ratio, 574
assumptions, 699 Bayesian inference, 550
completely randomized design (see Completely Bayesian point estimation
randomized design) Bayes’ rule, 552
F-test, 546 criteria for finding Bayesian estimate, 558–566
linear models, 546–547 likelihood function, 552–553
logarithmic transformation, 545 marginal distribution, 553
Minitab, 537–539 population proportion, 554–555
missing observations, 546 posterior distribution, 552
multiple comparisons, 528–535 probability distribution, 552–553
multiple regressions, 444–445 Bayes’ rule, 75b
one-way (see One-way ANOVA) Bell-shaped curve, 32
R code, 536–537 Bernoulli population, 246
regression, 430–432 Bernoulli random variable, 223, 250
SAS, 540–544 probability function of, 113
simple regression, 430–432 Best linear unbiased estimator (BLUE), 421–422
SPSS, 540 Beta-binomial distribution, 557
square root transformation, 545 Beta PDF, 399, 401f
treatments, 498–505 Binomial distribution, normal approximation,
two-way (see Two-way ANOVA) 206–208
Anderson–Darling test, 387–388 correction for continuity, 207b
Angular transformation, ANOVA, 545 Binomial experiment, 113
ANOVA. See Analysis of variance (ANOVA) Binomial formula, 224
Area sampling, 11 Binomial probability distribution, 113–117
Asymptotic properties, 303–304 Poisson approximation, 119b
Asymptotic relative efficiency, 304 Binomial random variables, 246
Average deviation, 30 expected value of, 128
Averaged squared errors (ASES), 305–306 mean of, 116b
Average weight loss estimation, 262 moment-generating function of, 116b
Axiomatic definition of probability, 57b variance of, 116b
Binomial theorem, 114
B Birthday problem, 108–109
Bar graph Bivariate data, 715–718
definition, 13 Bivariate probability distributions, 139–140
Pareto chart, 14–17, 15f Blinding, 465
Bayes, Thomas, 549 Blocking, 466
Bayes decision, 579 Bootstrap methods, 645–651
Bayesian confidence interval. See Credible intervals R code, 679
Bayesian decision theory SAS, 686
decision-making process, 576–577 Box plot, 33–35, 33b
statistical theory, 577 outliers, 695–697

791
792 Index

C two population parameters, 289–297

Cauchy distribution, 244 upper and lower confidence limits, 261–262
Central limit theorem (CLT), 164b, 262, 269–270, Conjugate prior, 557
271–272 Contingency table, chi-square tests
Chapman–Kolmogorov equation, 739 definition, 376–377
Chebyshev’s theorem, 160b independent factors, 377b
Chi-square distribution, 186–191, 284–285 Minitab, 407
degrees of freedom, 186 Continuity correction factor, 383–384
density, 285–286, 285f Control plot, Taguchi methods, 484, 484f
probabilities, 751t Correction factor, 506
probability density, 188f Correlation analysis
random variable, 133, 133b Fisher z-transform, 438
Chi-square tests independent variables, 436–437
contingency table, 376–379 maximum likelihood estimator, 437
multinomial distribution, 374–376 simple linear regression model, 436–437
Pearson’s, 381–384 Correlation coefficient, 147–149, 436–437
Cluster sampling, 11 Counting random variable, 117
Coefficient of determination, 419, 457 Covariance, 147–149
Complement set, 735, 735f Credible intervals
Completely randomized design conditional distribution, 569
ANOVA decomposition, 507–508, 507f definition, 568
between-groups variability, 505 posterior distribution, 569, 570f
correction factor, 506 Cross-sectional data, 6
decomposition of SS, 507, 507f Cumulative binomial probabilities, 747t
null hypothesis, 505 Cumulative distribution function (cdf), 84, 85
one-way ANOVA, 508b, 509b, 512–517 Cumulative probability distribution, 236
population means, 505
p-value approach, 510–511
SSE, 506 D
unbiased estimator, 508–509 Data
within-groups variability, 505 bivariate, 715–718
Composite hypothesis testing, 315 collection, 3, 3b
Computers and statistics, 39 cross-sectional, 6
Conditional probability graphical representation, 13–26 (see also
definition, 71 Graphical representation)
law of total probability, 74b nominal, 6–7
properties of, 71b numerical description, 26–38
Conditional probability distributions, 145 ordinal, 6–7
Confidence intervals quantitative, 5
computer examples, 300, 301 time series, 6
confidence coefficient, 262 transformation, 703–705
degrees of freedom, 263 types of, 5–7
interval estimation, 261 de Moivre, Abraham, 177
normal population, 264 Degrees of freedom, 186, 284–285
one sample, 269–284 Descriptive point estimates, 299
pivotal quantity, 263 Descriptive statistics, 5
population variance, 284–289 Design of experiments (DOE)
probability density, pivot, 265, 265f basic terminology, 461–465
proportion, 272 factorial design, 477–481
sample mean, 262–263 Minitab, 491
sampling distributions, 267, 308 optimal design, 481–483
Tukey’s method, 530b R code, 489–491
 Index 793

replication, randomization, and blocking, Finite set, 733

466–468 Finite variance, 259
sample size and power, 493 Fisher, Ronald Aylmer, 1
SAS, 491–493 Fisher z-transform, 438
specific designs, 469–476 Fractional factorial design, 480
Taguchi methods, 484 Friedman test, 775t
temperature effect, 493–494 Minitab, 630–632
Digamma function, 235–236 R code, 630
Discrete distribution, 256 treatment effects, 621
Discrete random variable, 117 Full conditionals, 674
Discrete uniform distribution, 376 Full factorial design, 480
Distribution-free tests, 695
income distribution of families, 590–591, 591f G
nonparametric confidence interval, 592–597 Galton, Francis, 409
nonparametric hypothesis tests Gamma probability distribution, 129–134, 224, 233
(see Nonparametric hypothesis tests) Gauss, Carl Friedrich, 111
outliers, 695 Gaussian distribution, 122
parametric tests, 590 Gaussian probability distribution, 381, 388
projects for, 635–637 Gauss–Markov theorem, 422b
Dobson units, 277–278 Geometric distribution, 229, 253
DOE. See Design of experiments (DOE) Gibbs algorithm (Gibbs sampler), 673–676
Dotplot, 689–690, 690f, 700f Goodness-of-fit tests
Double-blind treatment method, 465 Anderson–Darling test, 387–388
brain cancer, 397–398
E chi-square tests, 372–380, 381–384
Elementary statistics. See also Statistics contingency table, 376–379
course, 270 global warming, 392–393
Empirical distribution function, 307 hurricane Katrina, 393–396
Empty set (null set), 733 Kolmogorov–Smirnov test, 384–387
Equality of variances, 706–709 Minitab, 407
testing, 352b multinomial distribution, 372–373, 374–376
Ergodic theorem, 741 national unemployment, 396–397
Error probability distribution, 238–239 P-P plots, 389–391
Error variance estimation, 422 prostate cancer, 401–402
Estimation theory, 220 Q-Q plots, 389–391
Expectation maximization (EM) algorithm, rainfall, 399–400
651–662 R-commands, 406–407
R code, 682 Shapiro–Wilk test, 388–389
Experimental error, 465 Graphical representation
Exponential family of probability distributions, bar, 13
256–257 dotplot, 689–690, 690f, 700f
Exponential power, 238–239 frequency table, 17–18
Exponential probability distribution, 132 grouped data, 17–18
histogram, 18
F pie chart, 15, 16f
Factorial design quantile-quantile (QQ) plot, 691–693
fractional, 480 relative frequency, 17
full, 480 scatter plot, 690, 690f, 691
one-factor-at-a-time design, 478–480 side-by-side box plots, 690
Factorization criteria, joint sufficiency, 254b stem-and-leaf plot, 16, 17
F-distribution, 195–197 Greco–Latin square, 475
Finite population correction factor, 181 Grouped data, numerical measures, 30–32
794 Index

H Minitab, 631
Hardy–Weinberg law, 109, 114–115 R code, 629
Highest posterior density (HPD) interval, 571–572 SAS, 634
Histogram, 385, 385f, 700f SPSS, 633
of data, 704f
definition, 18 L
guideline, 18b Large sample approximations, 205–210
Homoscedasticity, 447 binomial distribution, normal approximation,
Hypothesis testing 206–208
composite, 315 Large sample confidence intervals
level of significance, 315 difference of two means, 290b
likelihood ratio tests, 328–333, 330b Minitab, 302
method for, 314b p1-p2, 269b
Neyman–Pearson Lemma, 323–328 Latin square design
p-value, 333–336 definition, 472
sample size, 320–321 Greco–Latin square, 475
simple, 315 R code, 490
single parameter, 333–344 Least-squares equations, 414–416
two samples, 345–359 Least-squares estimators
type I error, 315 definition, 414
type II error, 315 Gauss–Markov theorem, 447
inferences, 425–433
I properties of, 420–422
Independent variables, 462 Least-squares line, 414
Inferential statistics, 5 Least-squares, method of, 413–414
Infinite set, 733 Least-squares regression line, 412, 412f
Informative priors, 555–556 Least-squares regression model, 447–448
Interquartile range (IQR), 27–28 Level of significance, hypothesis testing, 315
Invariance property, 242 Likelihood ratio tests (LRT), 328–333, 330b
Limit theorems, 159–168
J central limit theorem (CLT), 164b
Jackknife method, 640–645 Chebyshev’s theorem, 160b
R code, 680 law of large numbers, 162b
SAS, 686 Linear regression models
Jeffreys’ hypothesis testing criterion, 574 ANOVA, 546–547
Joint density function, 251, 256 coefficient of determination, 457
Joint probability distributions, 139–152 correlation analysis, 436–440
bivariate distributions, 139–140 least-squares estimators, 425–433
conditional expectation, 145 matrix notation, 440–446
covariance and correlation, 147–149 Minitab, 453
marginal pmf, 141 outliers and high leverage points, 457–458
Joint probability mass function, 250 particular value prediction, 433–436
regression diagnostics, 446–449
K SAS, 454–456
Kolmogorov, Andrei Nikolaevich, 53 scatterplots, 456
Kolmogorov–Smirnov test, one sample test simple, 411–425
statistics, 783t SPSS, 454
Kronecker Delta function, 559 Logarithmic transformation, ANOVA, 545
Kruskal–Wallis test Log-likelihood function, 229–231
asymptotic distribution, 620 Loss function, Taguchi methods, 485, 485f
chi-square distribution, 619–620 Lower confidence limit, 261–262, 264,
description, 618 266, 267
 Index 795

M continuous case, 667b

Maclaurin’s expansion, with Poisson random discrete case, 666b
variable, 118 random-walk, 669
Marginal pmf/pdf, 141 Metropolis-Hastings (M-H) algorithm,
Margin of error and sample size, 272–275 670–672
Markov chain Monte Carlo (MCMC) methods, continuous case, 671b
662–678 discrete case, 670b
issues in, 676 Minimal sufficient statistics, 222
Metropolis algorithm, 666–669 Minimum variance unbiased estimator (MVUE),
R code, 680 240
Markov chains, 737 Minitab, 301–302
aperiodic, 740–741 ANOVA, 537–539
ergodic, 741 design of experiments, 491
homogeneous, 738 goodness-of-fit tests, 407
irreducible, 740 linear regression models, 453
periodic, 740 nonparametric tests, 630–632
random walk chain, 739 statistical estimation, 301–302
steady state, 741 t-test, 363
stochastic/random process, 737 Model
transient, 741 issues in, 712–719
transition probabilities, 737 for univariate data, 713–715
transition/stochastic matrix, 738 Moment-generating function (MGF)
Matrix notation of Bernoulli random variable, 113
independent observations, 441 of binomial random variable, 116b
least-squares estimators, 442 chi-square random variable, 133b
linear equations, 441 exponential random variable, 132b
multiple regression model, 441 gamma probability distribution, 130b
Maximum likelihood equations (MLE), 231, 233 moments and, 91–104
definition, 227 normal probability distribution, 123b
log-likelihood function, 229–231 of Poisson random variable, 118b
optimization, 234 properties of, 101b
parameter values, 227 uniform probability distribution, 121b
probability distributions, 234–240 Multifactor experiments, 464
Mean Multinomial distribution, 374–376
of binomial random variable, 116b Multiple comparisons, ANOVA
chi-square random variable, 133b studentized range distribution, 529–530
exponential random variable, 132b Tukey’s method, 530b
gamma probability distribution, 130b Multiple linear regression model
normal probability distribution, 123b ANOVA table, 445, 445t
of Poisson random variable, 118b definition, 411–412
uniform probability distribution, 121b
and variance estimation, 306 N
Mean square error (MSE), 248, 500 Negative binomial distribution, 242
Mean square treatment (MST), 500 Newton–Raphson iterative method, 306
Median test Neyman, Jerzy, 311
hypergeometric distribution, 610 Neyman–Fisher factorization theorem, 252b
hypothesis testing procedure, 610b Neyman–Pearson Lemma, 323–328
large sample, 611b Nightingale, Florence, 687
Minitab, 630–632 Noise, 463
sample median, 610, 610t Nominal data, 6–7
Method of moments, 222–227 Noninformative priors, 555–556
Metropolis algorithm Nonparametric analysis vs. parametric, 719–721
796 Index

Nonparametric confidence interval Optimal design

binomial distribution, 592 choice of optimal sample size, 482–483
central limit theorem, 592 sequential design, 481–482
ordered sample, 593, 593f simultaneous experiment design, 481–482
population median, 593 Optimization, 234
Nonparametric hypothesis tests Order statistics, 200–205
for one sample, 597–609 Ordinal data, 6–7
for two samples, 609–618 Orthogonal Latin squares, 475
Nonparametric tests. See Distribution-free tests Outliers
Normal distribution, 225 box plot, 695–697
Normality, assumption, 699–700 distribution-free test, 695
Normal probability distribution, 122–129 and high leverage points, 457–458
Normal probability plots, 701, 701f, 702f, 704f, modified z-score, 695
705f, 723f, 724f value, 694
for ANOVA, 514f, 545 z-score, 695
Nuisance variables, 462
Null hypothesis, 312–313 P
Numerical description, data Paired comparison tests, 606–607
average deviation, 30 Parametric analysis
bell-shaped curve, 32 brain cancer, 397–398
empirical rule, 32b global warming, 392–393
grouped data, numerical measures, 30–32 hurricane Katrina, 393–396
interquartile range, 27–28 national unemployment, 396–397
lower quartile, 27 vs. nonparametric analysis, 719–721
median, 27 prostate cancer, 401–402
sample mean (empirical mean), 26 rainfall, 399–400
sample standard deviation, 26 Pareto chart, 14–17
sample variance, 26 Pareto distribution, 244
upper quartile, 27 Pearson, Karl, 371
Numerical unbiasedness and consistency, 305 Pearson’s chi-square tests
cumulative probability distribution, 381
O Gaussian probability distribution, 381
Observables Percentage point of F-distributions, 752t
for Bayesian decision theory, 576–583 Pie chart, 15, 16f
definition, 579 Pivotal method. see Confidence intervals
predicting future, 587–588 Placebo, 465
Observational experiment, 463 Point estimators
One-factor-at-a-time design, 478–480 method of maximum likelihood, 227–234
One-parameter Weibull distribution, 260 method of moments, 222–227
One sample confidence intervals sufficiency, 250–258
large sample, 269–271 unbiased estimators, 245–249
proportion, 272 Poisson, Siméon-Denis, 117
small sample, 275–278 Poisson distribution, 226, 230, 260
One-tailed test, 313–314 Poisson probability distribution, 117–119
One-way ANOVA, 465 discrete random variable and, 117
assumption testing, 512–517 Poisson random variables, 227
k32 populations, 508b, 509b definition of, 117
Minitab, 537–539 mean, 118b
model for, 517 moment-generating function of, 118b
R code, 513 variance, 118b
SAS, 540 Pooled sample variance, 291
SPSS, 540 Pooled t-test, 347
 Index 797

Population sample space, 55b

defined, 4 special distribution functions, 112–139
standard deviation, 274–275 trial, 55
Population parameters, confidence interval p-value
large sample, 290b approach, 510–511
small sample, 290b hypothesis testing, 333–336
Population variance, confidence interval
chi-square density, 285–286, 285f Q
chi-square distribution, 284–285 Quadratic loss function, 486, 486f, 559
Posterior distribution Quantile-quantile (QQ) plot, 389–391, 691–693
Bayesian point estimation, 552–568 Quantitative data, 5
definition, 552
Posterior mean, 559 R
Posterior odds ratio, 574 Random assignment procedure, 468
Power transformation, 715 Randomization, 467
P-P plots, 389–391 Randomized complete block design.
Prediction interval, normal population, 309–310 See also Two-way ANOVA
Prior odds ratio, 574 definition, 469
Probabilistic model. See Statistics R code, 490
Probability density, 241, 265, 265f replications, 470b
Probability distribution, 381–392 SAS, 491
common, 743 Randomness test
Probability distribution function (PDF), 112–139, asymptotic normal distribution, 635
220, 221–222, 234–240 Minitab, 637
references for, 112 nonparametric procedure, 635
Probability function (pf), Bernoulli random Random variables
variable, 113 Bernoulli (see Bernoulli random variable)
Probability mass function, 221–222, 230 binomial (see Binomial random variables)
Probability tables counting, 117
chi-square probabilities, 751t and probability distributions, 82–91
cumulative binomial probabilities, 747t Random variables functions, 152–159
Friedman test, 775t distribution functions method, 152–153
Kolmogorov–Smirnov one sample test statistics, functions of, 155
783t pdf, 154
percentage point of F-distributions, 752t probability integral transformation,
standard norms table, 749t 154–155
studentized range q table, 781t transformation method, 155–158
t-table, 750t Random-walk metropolis, 669
Wilcoxon rank sum test, 767t Rao, C.R., 219
Wilcoxon signed rank test, 762t Rao–Blackwell Theorem, 258b
Probability theory Rayleigh distribution, 239, 239f, 261
axiomatic definition, 57b R code
concept of, 54 Bayesian estimation inference, 584–587
counting techniques and calculation of, 63–70 design of experiments, 489–491
experiment, defined, 55 goodness-of-fit tests, 289–297
impossible event, 55b linear regression models, 450–453
mutually exclusive/ disjoint, 55 nonparametric tests, 628–630
origin of, 54 one-way ANOVA, 513
probability, defined, 55, 57 two-way ANOVA, 536
probability space, 55b Regression diagnostics, 446–449
properties of, 58b Rejection region (Critical region), 313
sample point, 55b Relative frequency, 17, 18
798 Index

Replication one-to-one correspondence, 736

definition, 466 properties, 735–736
procedure for randomized complete block design, set, defined, 733
470b subset, 733
Response variable, 461–462 symmetric difference, 735–736
R language, 745 union, 734, 734f
Robust estimation, 304–305 universal set, 733
Venn diagram, 733, 734f
S Shapiro–Wilk test, 388–389
Sampling Shortest length confidence interval, 264
area, 11 Side-by-side box plots, 690
biased, 8 one-way ANOVA, 512–516, 514f
chi-square distribution, 186–191 Sign test
cluster, 11 binomial distribution, 598
defined, 4 hypothesis testing procedure, 598b
distribution, 179 large random sample, 600b
errors in, 12 Minitab, 630
F-distribution, 195–197 null hypothesis testing, 597
finite population correction factor, 181 population distribution, 597–598
large sample approximations, 205–210 R code, 628
Minitab, 213–214 z-transform, 600
multiphase, 11 Simple hypothesis testing, 315
order statistics, 200–205 Simple linear regression models
population distribution, 184–200 definition, 412
R code, 211–213 derivation of b0 and b1, 414–418
representative, 8 error variance estimation, 422
sample, defined, 178 least-squares estimators, 420–422
SAS, 214–215 least-squares regression line, 412, 412f
simple random, 8 method of least-squares, 413–414
size, 12–13 quality of regression, 418–419
SPSS, 214 scatter diagram, 411, 411f
standard error, 180 Simple random sampling
stratified, 9, 10, 10b advantages, 9b
student t-distribution, 191–195 definition, 8
systematic, 9 Simple regression line, 417, 417f
SAS, 303 Single-factor experiments, 464
ANOVA, 540–544 Skewness and Kurtosis, 96–101, 701
design of experiments, 491–493 Small sample confidence intervals, 307
linear regression models, 454–456 difference of two means, 291b
Nonparametric tests, 633–634 Minitab, 302
statistical estimation, 303 simulation of coverage, 307
Scatter diagram, 287, 411, 411f Smith-Satterthwaite procedure, 348–350
Scatter plot, 412, 412f, 456, 690, 690f, 691 SPSS
Set theory ANOVA, 540
complement, 735, 735f linear regression models, 454
countably infinite, 736 Nonparametric tests, 632–633
difference, 735–736 statistical estimation, 302–303
disjoint/mutually exclusive, 735 Squared error loss function, 559
elements/members, 733 Square root transformation, ANOVA, 545
empty set (null set), 733 Standard error, 180
finite, 733 Standard normal density, 257
infinite, 733 Standard normal random variable, 122–123
intersection, 734, 734f Standard norms table, 749t
 Index 799

Standard pivotal quantity, 263 Transformation(s)

Statistic(s) for ANOVA, 545
concepts of, 4 power, 715
descriptive, 5 Transition probabilities, 737
inferential, 5 function, 665
population, 4 n-step, 739
sampling, 4 positive transition matrix, 740
Statistical decision, 312 Treatment variables, 462
making, 577–578 Truncated exponential distribution, 243
Statistical hypotheses, 312 t-table, 750t
Stem-and-leaf plot, 16, 17 t-test
Sticker price, 726f, 727f, 728f assumptions, 699
Stratified sample Minitab, 363
definition, 9 one-sample, 360, 362
selection procedure, 10b paired samples, 363
uses, 11b pooled, 347
Studentized range distribution, 529–530 SAS, 366–368
Studentized range q table, 781t SPSS, 365–366
Student t-distribution, 191–195, 284–285 two sample, 362
Subjective probability, 550–551 Tukey, John W., 495
Subset, 733 Tukey–Kramer method, 533
proper subset, 733 Tukey’s method
Sufficient estimator calculations of, 530, 530t
conditional probability, 252 confidence intervals, 529
definition, 250 Minitab, 539
density functions, 256–257 R code, 537
factorization criterion, 253, 254b SAS, 541
Neyman–Fisher factorization criteria, 252b SPSS, 540
Rao–Blackwell theorem, 258b Two random samples, hypothesis testing, 345–359
Sum of squares of errors (SSE), 499, 506 dependent samples, 353–356
Systematic sampling independent samples, 345–353
definition, 9 Two-way ANOVA, 465
selection procedure, 9b computational procedure for, 524b
Minitab, 538
nonrandom effect, 521
T null hypothesis, 523
Taguchi, Genichi, 459 R code, 536
Taguchi methods step-by-step computational procedure, 523–525
control plot, 484, 484f sums of squares, 522
design parameters, 487 two-way classification, 521–522, 522t
engineering designs, 484 unbiased estimator, 523
goal post mentality, 485 Type I error, hypothesis testing, 315
loss function, 485, 485f Type II error, hypothesis testing, 315
quadratic loss function, 486, 486f
quality control, 484 U
Test of independence, 709–710 Ulam, Stanislaw, 639
Test statistics (TS), 313 Unbiased estimators
Three-parameter gamma PDF, 234, 235, definition, 245
396, 397f mean square error, 248
Three-parameter Weibull PDF, 398, 398f sample mean, 246
Time series data, 6 variance, 246–247
Time to failure and/or time between failure (TBF), Uniform maximum likelihood estimation, 299
721–731 Uniform probability distribution, 120–122, 231, 231f
800 Index

Univariate data, 713–715 Weibull PDF, 234, 236, 238, 238f, 394, 394f,
Upper confidence limit, 261–262, 264, 266, 267 401–402, 402f
Wilcoxon rank sum test, 767t
V hypothesis testing procedure, 613b
large sample, 615b
Variance
R code, 629
of binomial random variable, 116b
SAS, 634
chi-square random variable, 133b
SPSS, 632
exponential random variable, 132b
Wilcoxon signed rank test, 762t
gamma probability distribution, 130b
hypothesis testing procedure, 601b
normal probability distribution, 123b
large samples, 604b
of Poisson random variable, 118b
Minitab, 630–632
uniform probability distribution, 121b
R code, 628
Venn diagram, 733, 734f
Wilcoxon tests vs. normal approximation, 635
Wolfowitz, Jacob, 589
W
Wald–Wolfowitz test. See Randomness test Z
Weibull cumulative probability distribution, 395, z-score test, 695
395f, 401–402, 402f Z-transform, 270