-
Notifications
You must be signed in to change notification settings - Fork 0
/
point_estimation.html
1164 lines (1132 loc) · 71.9 KB
/
point_estimation.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"><head>
<meta charset="utf-8">
<meta name="generator" content="quarto-1.6.1">
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
<title>19 Point Estimation – Resampling statistics</title>
<style>
code{white-space: pre-wrap;}
span.smallcaps{font-variant: small-caps;}
div.columns{display: flex; gap: min(4vw, 1.5em);}
div.column{flex: auto; overflow-x: auto;}
div.hanging-indent{margin-left: 1.5em; text-indent: -1.5em;}
ul.task-list{list-style: none;}
ul.task-list li input[type="checkbox"] {
width: 0.8em;
margin: 0 0.8em 0.2em -1em; /* quarto-specific, see https://github.com/quarto-dev/quarto-cli/issues/4556 */
vertical-align: middle;
}
/* CSS for citations */
div.csl-bib-body { }
div.csl-entry {
clear: both;
margin-bottom: 0em;
}
.hanging-indent div.csl-entry {
margin-left:2em;
text-indent:-2em;
}
div.csl-left-margin {
min-width:2em;
float:left;
}
div.csl-right-inline {
margin-left:2em;
padding-left:1em;
}
div.csl-indent {
margin-left: 2em;
}</style>
<script src="site_libs/quarto-nav/quarto-nav.js"></script>
<script src="site_libs/quarto-nav/headroom.min.js"></script>
<script src="site_libs/clipboard/clipboard.min.js"></script>
<script src="site_libs/quarto-search/autocomplete.umd.js"></script>
<script src="site_libs/quarto-search/fuse.min.js"></script>
<script src="site_libs/quarto-search/quarto-search.js"></script>
<meta name="quarto:offset" content="./">
<link href="./framing_questions.html" rel="next">
<link href="./inference_intro.html" rel="prev">
<script src="site_libs/quarto-html/quarto.js"></script>
<script src="site_libs/quarto-html/popper.min.js"></script>
<script src="site_libs/quarto-html/tippy.umd.min.js"></script>
<script src="site_libs/quarto-html/anchor.min.js"></script>
<link href="site_libs/quarto-html/tippy.css" rel="stylesheet">
<link href="site_libs/quarto-html/quarto-syntax-highlighting.css" rel="stylesheet" id="quarto-text-highlighting-styles">
<script src="site_libs/bootstrap/bootstrap.min.js"></script>
<link href="site_libs/bootstrap/bootstrap-icons.css" rel="stylesheet">
<link href="site_libs/bootstrap/bootstrap.min.css" rel="stylesheet" id="quarto-bootstrap" data-mode="light">
<script id="quarto-search-options" type="application/json">{
"location": "sidebar",
"copy-button": false,
"collapse-after": 3,
"panel-placement": "start",
"type": "textbox",
"limit": 50,
"keyboard-shortcut": [
"f",
"/",
"s"
],
"show-item-context": false,
"language": {
"search-no-results-text": "No results",
"search-matching-documents-text": "matching documents",
"search-copy-link-title": "Copy link to search",
"search-hide-matches-text": "Hide additional matches",
"search-more-match-text": "more match in this document",
"search-more-matches-text": "more matches in this document",
"search-clear-button-title": "Clear",
"search-text-placeholder": "",
"search-detached-cancel-button-title": "Cancel",
"search-submit-button-title": "Submit",
"search-label": "Search"
}
}</script>
<script type="text/javascript">
$(document).ready(function() {
$("table").addClass('lightable-paper lightable-striped lightable-hover')
});
</script>
<script src="https://cdnjs.cloudflare.com/polyfill/v3/polyfill.min.js?features=es6"></script>
<script src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-chtml-full.js" type="text/javascript"></script>
<script type="text/javascript">
const typesetMath = (el) => {
if (window.MathJax) {
// MathJax Typeset
window.MathJax.typeset([el]);
} else if (window.katex) {
// KaTeX Render
var mathElements = el.getElementsByClassName("math");
var macros = [];
for (var i = 0; i < mathElements.length; i++) {
var texText = mathElements[i].firstChild;
if (mathElements[i].tagName == "SPAN") {
window.katex.render(texText.data, mathElements[i], {
displayMode: mathElements[i].classList.contains('display'),
throwOnError: false,
macros: macros,
fleqn: false
});
}
}
}
}
window.Quarto = {
typesetMath
};
</script>
<link rel="stylesheet" href="style.css">
<link rel="stylesheet" href="font-awesome.min.css">
</head>
<body class="nav-sidebar floating">
<div id="quarto-search-results"></div>
<header id="quarto-header" class="headroom fixed-top">
<nav class="quarto-secondary-nav">
<div class="container-fluid d-flex">
<button type="button" class="quarto-btn-toggle btn" data-bs-toggle="collapse" role="button" data-bs-target=".quarto-sidebar-collapse-item" aria-controls="quarto-sidebar" aria-expanded="false" aria-label="Toggle sidebar navigation" onclick="if (window.quartoToggleHeadroom) { window.quartoToggleHeadroom(); }">
<i class="bi bi-layout-text-sidebar-reverse"></i>
</button>
<nav class="quarto-page-breadcrumbs" aria-label="breadcrumb"><ol class="breadcrumb"><li class="breadcrumb-item"><a href="./point_estimation.html"><span class="chapter-number">19</span> <span class="chapter-title">Point Estimation</span></a></li></ol></nav>
<a class="flex-grow-1" role="navigation" data-bs-toggle="collapse" data-bs-target=".quarto-sidebar-collapse-item" aria-controls="quarto-sidebar" aria-expanded="false" aria-label="Toggle sidebar navigation" onclick="if (window.quartoToggleHeadroom) { window.quartoToggleHeadroom(); }">
</a>
<button type="button" class="btn quarto-search-button" aria-label="Search" onclick="window.quartoOpenSearch();">
<i class="bi bi-search"></i>
</button>
</div>
</nav>
</header>
<!-- content -->
<div id="quarto-content" class="quarto-container page-columns page-rows-contents page-layout-article">
<!-- sidebar -->
<nav id="quarto-sidebar" class="sidebar collapse collapse-horizontal quarto-sidebar-collapse-item sidebar-navigation floating overflow-auto">
<div class="pt-lg-2 mt-2 text-left sidebar-header">
<div class="sidebar-title mb-0 py-0">
<a href="./">Resampling statistics</a>
</div>
</div>
<div class="mt-2 flex-shrink-0 align-items-center">
<div class="sidebar-search">
<div id="quarto-search" class="" title="Search"></div>
</div>
</div>
<div class="sidebar-menu-container">
<ul class="list-unstyled mt-1">
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./index.html" class="sidebar-item-text sidebar-link">
<span class="menu-text">Python version</span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./preface_third.html" class="sidebar-item-text sidebar-link">
<span class="menu-text">Preface to the third edition</span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./preface_second.html" class="sidebar-item-text sidebar-link">
<span class="menu-text">Preface to the second edition</span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./intro.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">1</span> <span class="chapter-title">Introduction</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./resampling_method.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">2</span> <span class="chapter-title">The resampling method</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./what_is_probability.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">3</span> <span class="chapter-title">What is probability?</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./about_technology.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">4</span> <span class="chapter-title">Introducing Python and the Jupyter notebook</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./resampling_with_code.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">5</span> <span class="chapter-title">Resampling with code</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./resampling_with_code2.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">6</span> <span class="chapter-title">More resampling with code</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./sampling_tools.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">7</span> <span class="chapter-title">Tools for samples and sampling</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./probability_theory_1a.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">8</span> <span class="chapter-title">Probability Theory, Part 1</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./probability_theory_1b.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">9</span> <span class="chapter-title">Probability Theory Part I (continued)</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./more_sampling_tools.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">10</span> <span class="chapter-title">Two puzzles and more tools</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./probability_theory_2_compound.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">11</span> <span class="chapter-title">Probability Theory, Part 2: Compound Probability</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./probability_theory_3.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">12</span> <span class="chapter-title">Probability Theory, Part 3</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./probability_theory_4_finite.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">13</span> <span class="chapter-title">Probability Theory, Part 4: Estimating Probabilities from Finite Universes</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./sampling_variability.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">14</span> <span class="chapter-title">On Variability in Sampling</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./monte_carlo.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">15</span> <span class="chapter-title">The Procedures of Monte Carlo Simulation (and Resampling)</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./standard_scores.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">16</span> <span class="chapter-title">Ranks, Quantiles and Standard Scores</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./inference_ideas.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">17</span> <span class="chapter-title">The Basic Ideas in Statistical Inference</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./inference_intro.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">18</span> <span class="chapter-title">Introduction to Statistical Inference</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./point_estimation.html" class="sidebar-item-text sidebar-link active">
<span class="menu-text"><span class="chapter-number">19</span> <span class="chapter-title">Point Estimation</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./framing_questions.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">20</span> <span class="chapter-title">Framing Statistical Questions</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./testing_counts_1.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">21</span> <span class="chapter-title">Hypothesis-Testing with Counted Data, Part 1</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./significance.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">22</span> <span class="chapter-title">The Concept of Statistical Significance in Testing Hypotheses</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./testing_counts_2.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">23</span> <span class="chapter-title">The Statistics of Hypothesis-Testing with Counted Data, Part 2</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./testing_measured.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">24</span> <span class="chapter-title">The Statistics of Hypothesis-Testing With Measured Data</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./testing_procedures.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">25</span> <span class="chapter-title">General Procedures for Testing Hypotheses</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./confidence_1.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">26</span> <span class="chapter-title">Confidence Intervals, Part 1: Assessing the Accuracy of Samples</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./confidence_2.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">27</span> <span class="chapter-title">Confidence Intervals, Part 2: The Two Approaches to Estimating Confidence Intervals</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./reliability_average.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">28</span> <span class="chapter-title">Some Last Words About the Reliability of Sample Averages</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./correlation_causation.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">29</span> <span class="chapter-title">Correlation and Causation</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./how_big_sample.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">30</span> <span class="chapter-title">How Large a Sample?</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./bayes_simulation.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">31</span> <span class="chapter-title">Bayesian Analysis by Simulation</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./references.html" class="sidebar-item-text sidebar-link">
<span class="menu-text">References</span></a>
</div>
</li>
<li class="sidebar-item sidebar-item-section">
<div class="sidebar-item-container">
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-1" role="navigation" aria-expanded="true">
<span class="menu-text">Appendices</span></a>
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-1" role="navigation" aria-expanded="true" aria-label="Toggle section">
<i class="bi bi-chevron-right ms-2"></i>
</a>
</div>
<ul id="quarto-sidebar-section-1" class="collapse list-unstyled sidebar-section depth1 show">
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./exercise_solutions.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">A</span> <span class="chapter-title">Exercise Solutions</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./technical_note.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">B</span> <span class="chapter-title">Technical Note to the Professional Reader</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./acknowlegements.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">C</span> <span class="chapter-title">Acknowledgements</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./code_topics.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">D</span> <span class="chapter-title">Code topics</span></span></a>
</div>
</li>
<li class="sidebar-item">
<div class="sidebar-item-container">
<a href="./errors_suggestions.html" class="sidebar-item-text sidebar-link">
<span class="menu-text"><span class="chapter-number">E</span> <span class="chapter-title">Errors and suggestions</span></span></a>
</div>
</li>
</ul>
</li>
</ul>
</div>
</nav>
<div id="quarto-sidebar-glass" class="quarto-sidebar-collapse-item" data-bs-toggle="collapse" data-bs-target=".quarto-sidebar-collapse-item"></div>
<!-- margin-sidebar -->
<div id="quarto-margin-sidebar" class="sidebar margin-sidebar">
<nav id="TOC" role="doc-toc" class="toc-active">
<h2 id="toc-title">Table of contents</h2>
<ul>
<li><a href="#ways-to-estimate-the-mean" id="toc-ways-to-estimate-the-mean" class="nav-link active" data-scroll-target="#ways-to-estimate-the-mean"><span class="header-section-number">19.1</span> Ways to estimate the mean</a>
<ul class="collapse">
<li><a href="#the-method-of-moments" id="toc-the-method-of-moments" class="nav-link" data-scroll-target="#the-method-of-moments"><span class="header-section-number">19.1.1</span> The Method of Moments</a></li>
<li><a href="#expected-value-and-the-method-of-moments" id="toc-expected-value-and-the-method-of-moments" class="nav-link" data-scroll-target="#expected-value-and-the-method-of-moments"><span class="header-section-number">19.1.2</span> Expected Value and the Method of Moments</a></li>
<li><a href="#the-maximum-likelihood-principle" id="toc-the-maximum-likelihood-principle" class="nav-link" data-scroll-target="#the-maximum-likelihood-principle"><span class="header-section-number">19.1.3</span> The Maximum Likelihood Principle</a></li>
</ul></li>
<li><a href="#choice-of-estimation-method" id="toc-choice-of-estimation-method" class="nav-link" data-scroll-target="#choice-of-estimation-method"><span class="header-section-number">19.2</span> Choice of Estimation Method</a></li>
<li><a href="#criteria-of-estimates" id="toc-criteria-of-estimates" class="nav-link" data-scroll-target="#criteria-of-estimates"><span class="header-section-number">19.3</span> Criteria of estimates</a>
<ul class="collapse">
<li><a href="#unbiasedness" id="toc-unbiasedness" class="nav-link" data-scroll-target="#unbiasedness"><span class="header-section-number">19.3.1</span> Unbiasedness</a></li>
<li><a href="#efficiency" id="toc-efficiency" class="nav-link" data-scroll-target="#efficiency"><span class="header-section-number">19.3.2</span> Efficiency</a></li>
<li><a href="#maximum-likelihood" id="toc-maximum-likelihood" class="nav-link" data-scroll-target="#maximum-likelihood"><span class="header-section-number">19.3.3</span> Maximum Likelihood</a></li>
</ul></li>
<li><a href="#criteria-of-the-criteria" id="toc-criteria-of-the-criteria" class="nav-link" data-scroll-target="#criteria-of-the-criteria"><span class="header-section-number">19.4</span> Criteria of the Criteria</a></li>
<li><a href="#estimation-of-accuracy-of-the-point-estimate" id="toc-estimation-of-accuracy-of-the-point-estimate" class="nav-link" data-scroll-target="#estimation-of-accuracy-of-the-point-estimate"><span class="header-section-number">19.5</span> Estimation of accuracy of the point estimate</a></li>
<li><a href="#sec-uses-of-mean" id="toc-sec-uses-of-mean" class="nav-link" data-scroll-target="#sec-uses-of-mean"><span class="header-section-number">19.6</span> Uses of the mean</a></li>
<li><a href="#conclusion" id="toc-conclusion" class="nav-link" data-scroll-target="#conclusion"><span class="header-section-number">19.7</span> Conclusion</a></li>
</ul>
</nav>
</div>
<!-- main -->
<main class="content" id="quarto-document-content">
<header id="title-block-header" class="quarto-title-block default">
<div class="quarto-title">
<h1 class="title"><span id="sec-point-estimation" class="quarto-section-identifier"><span class="chapter-number">19</span> <span class="chapter-title">Point Estimation</span></span></h1>
</div>
<div class="quarto-title-meta">
</div>
</header>
<p>One of the great questions in statistical inference is: How big is it? This can mean — How long? How deep? How much time? At what angle?</p>
<p>This question about size may pertain to a single object, of which there are many measurements; an example is the location of a star in the heavens. Or the question may pertain to a varied set of elements and their measurements; examples include the effect of treatment with a given drug, and the incomes of the people of the United States in 1994.</p>
<p>From where the observer stands, having only the evidence of a sample in hand, it often is impossible to determine whether the data represent multiple observations of a single object, or single (or multiple) observations of multiple objects. For example, from crude measurements of weight you could not know whether one person is being weighed repeatedly, or several people have been weighed once. Hence all the following discussion of point estimation is the same for both of these situations.</p>
<p>The word “big” in the first sentence above is purposely vague, because there are many possible kinds of estimates that one might wish to make concerning a given object or collection. For a single object like a star, one surely will wish to make a best guess about its location. But about the effects of a drug treatment, or the incomes of a nation, there are many questions that one may wish to answer. The <em>average</em> effect or income is a frequent and important object of our interest. But one may also wish to know about the amount of <em>dispersion</em> in the distribution of treatment effects, or of incomes, or the symmetry of the distribution. And there are still other questions one may wish to answer.</p>
<p>Even if we focus on the average, the issue often is less clear cut than we may think at first. If we are to choose a single number to characterize the population (universe) from which a given set of data has been drawn, what should that representative number be for the case at hand? The answer must depend on the purpose with which we ask the question, of course. There are several main possibilities such as the mean, the median, and the mode.</p>
<p>Even if we confine our attention to the mean as our measure of the central tendency of a distribution, there are various ways of estimating it, each of them having a different rationale. The various methods of estimation often lead to the same estimate, especially if the distribution is symmetric (such as the distribution of errors you make in throwing darts at a dart board). But in an asymmetric case such as a distribution of incomes, the results may differ among the contending modes of estimation. So the entire topic is more messy than appears at first look. Though we will not inquire into the complexities, it is important that you understand that the matter is not as simple as it may seem. (See <span class="citation" data-cites="savage1972foundations">Savage (<a href="references.html#ref-savage1972foundations" role="doc-biblioref">1972</a>)</span>, Chapter 15, for more discussion of this topic.)</p>
<section id="ways-to-estimate-the-mean" class="level2" data-number="19.1">
<h2 data-number="19.1" class="anchored" data-anchor-id="ways-to-estimate-the-mean"><span class="header-section-number">19.1</span> Ways to estimate the mean</h2>
<section id="the-method-of-moments" class="level3" data-number="19.1.1">
<h3 data-number="19.1.1" class="anchored" data-anchor-id="the-method-of-moments"><span class="header-section-number">19.1.1</span> The Method of Moments</h3>
<p>Since elementary school you have been taught to estimate the mean of a universe (or calculate the mean of a sample) by taking a simple arithmetic average. A fancy name for that process is “the method of moments.” It is the equivalent of estimating the center of gravity of a pole by finding the place where it will balance on your finger. If the pole has the same size and density all along its length, that balance point will be halfway between the endpoints, and the point may be thought of as the arithmetic average of the distances from the balance point of all the one-centimeter segments of the pole.</p>
<p>Consider this example:</p>
<p><strong>Example: Twenty-nine Out of Fifty People Polled Say They Will Vote For The Democrat. Who Will Win The Election? The Relationship Between The Sample Proportion and The Population Proportion in a Two-Outcome Universe.</strong></p>
<p>You take a random sample of 50 people in Maryland and ask which party’s candidate for governor they will vote for. Twenty-nine say they will vote for the Democrat. Let’s say it is reasonable to assume in this case that people will vote exactly as they say they will. The statistical question then facing you is: What proportion of the voters in Maryland will vote for the Democrat in the general election?</p>
<p>Your intuitive best guess is that the proportion of the “universe” — which is composed of voters in the general election, in this case — will be the same as the proportion of the sample. That is, 58 percent = 29/50 is likely to be your guess about the proportion that will vote Democratic. Of course, your estimate may be too high or too low in this particular case, but in the long run — that is, if you take many samples like this one — on the average the sample mean will equal the universe (population) proportion, for reasons to be discussed later.</p>
<p>The sample mean seems to be the “natural” estimator of the population mean in this and many other cases. That is, it seems quite natural to say that the best estimate is the sample mean, and indeed it probably is best. But why? This is the problem of inverse probability that has bedeviled statisticians for two centuries.</p>
<p>If the only information that you have (or that seems relevant) is the evidence of the sample, then there would seem to be no basis for judging that the shape and location of the population differs to the “left” or “right” from that of the sample. That is often a strong argument.</p>
<p>Another way of saying much the same thing: If a sample has been drawn randomly, each single observation is a representative estimator of the mean; if you only have one observation, that observation is your best guess about the center of the distribution (if you have no reason to believe that the distribution of the population is peculiar — such as not being symmetrical). And therefore the sum of 2, 3…n of such observations (divided by their number) should have that same property, based on basic principles.</p>
<p>But if you are on a ship at sea and a leaf comes raining down from the sky, your best guess about the location of the tree from which it comes is not directly above you, and if two leaves fall, the midpoint of them is not the best location guess, either; you know that trees don’t grow at sea, and birds sometimes carry leaves out to sea.</p>
<p>We’ll return to this subject when we discuss criteria of methods.</p>
</section>
<section id="expected-value-and-the-method-of-moments" class="level3" data-number="19.1.2">
<h3 data-number="19.1.2" class="anchored" data-anchor-id="expected-value-and-the-method-of-moments"><span class="header-section-number">19.1.2</span> Expected Value and the Method of Moments</h3>
<p>Consider this gamble: You and another person roll a die. If it falls with the “6” upwards you get $4, and otherwise you pay $1. If you play 120 times, at the end of the day you would expect to have (20 * $4 - 100 * $1 =) -$20 dollars. We say that -$20 is your “expected value,” and your expected value per roll is (-$20 / 120 =) $.166 or the loss of 1/6 of a dollar. If you get $5 instead of $4, your expected value is $0.</p>
<p>This is exactly the same idea as the method of moments, and we even use the same term — “expected value,” or “expectation” — for the outcome of a calculation of the mean of a distribution. We say that the expected value for the success of rolling a “6” with a single cast of a die is 1/6, and that the expected value of rolling a “6” or a “5” is (1/6 + 1/6 = ) 2/6.</p>
</section>
<section id="the-maximum-likelihood-principle" class="level3" data-number="19.1.3">
<h3 data-number="19.1.3" class="anchored" data-anchor-id="the-maximum-likelihood-principle"><span class="header-section-number">19.1.3</span> The Maximum Likelihood Principle</h3>
<p>Another way of thinking about estimation of the population mean asks: Which population(s) would, among the possible populations, have the highest probability of producing the observed sample? This criterion frequently produces the same answer as the method of moments, but in some situations the estimates differ. Furthermore, the logic of the maximum-likelihood principle is important.</p>
<p>Consider that you draw without replacement six balls — 2 black and 4 white — from a bucket that contains twenty balls. What would you guess is the composition of the bucket from which they were drawn? Is it likely that those balls came from a bucket with 4 white and 16 black balls? Rather obviously not, because it would be most unusual to get all the 4 white balls in your draw. Indeed, we can estimate the probability of that happening with simulation or formula to be about .003.</p>
<p>How about a bucket with 2 black and 18 whites? The probability is much higher than with the previous bucket, but it still is low — about .075.</p>
<p>Let us now estimate the probabilities for all buckets across the range of probabilities. In <a href="#fig-white_balls_universe" class="quarto-xref">Figure <span>19.1</span></a> we see that the bucket with the highest probability of producing the observed sample has the same proportions of black and white balls as does the sample. This is called the “maximum likelihood universe.” Nor should this be very surprising, because that universe obviously has an equal chance of producing samples with proportions below and above that observed proportion — as was discussed in connection with the method of moments.</p>
<p>We should note, however, that the probability that even such a maximum-likelihood universe would produce <em>exactly</em> the observed sample is very low (though it has an even lower probability of producing any <em>other</em> sample).</p>
<div class="cell" data-layout-align="center">
<div class="cell-output-display">
<div id="fig-white_balls_universe" class="quarto-float quarto-figure quarto-figure-center anchored" data-fig-align="center">
<figure class="quarto-float quarto-float-fig figure">
<div aria-describedby="fig-white_balls_universe-caption-0ceaefa1-69ba-4598-a22c-09a6ac19f8ca">
<img src="diagrams/white_balls_universe.svg" class="img-fluid quarto-figure quarto-figure-center figure-img" style="width:70.0%">
</div>
<figcaption class="quarto-float-caption-bottom quarto-float-caption quarto-float-fig" id="fig-white_balls_universe-caption-0ceaefa1-69ba-4598-a22c-09a6ac19f8ca">
Figure 19.1: Number of White Balls in the Universe (N=20)
</figcaption>
</figure>
</div>
</div>
</div>
</section>
</section>
<section id="choice-of-estimation-method" class="level2" data-number="19.2">
<h2 data-number="19.2" class="anchored" data-anchor-id="choice-of-estimation-method"><span class="header-section-number">19.2</span> Choice of Estimation Method</h2>
<p>When should you base your estimate on the method of moments, or of maximum likelihood, or still some other principle? There is no general answer. Sound estimation requires that you think long and hard about the purpose of your estimation, and fit the method to the purpose. I am well aware that this is a very vague statement. But though it may be an uncomfortable idea to live with, guidance to sound statistical method must be vague because it requires sound judgment and deep knowledge of the particular set of facts about the situation at hand.</p>
</section>
<section id="criteria-of-estimates" class="level2" data-number="19.3">
<h2 data-number="19.3" class="anchored" data-anchor-id="criteria-of-estimates"><span class="header-section-number">19.3</span> Criteria of estimates</h2>
<p>How should one judge the soundness of the process that produces an estimate? General criteria include <em>representativeness</em> and <em>accuracy</em>. But these are pretty vague; we’ll have to get more specific.</p>
<section id="unbiasedness" class="level3" data-number="19.3.1">
<h3 data-number="19.3.1" class="anchored" data-anchor-id="unbiasedness"><span class="header-section-number">19.3.1</span> Unbiasedness</h3>
<p>Concerning representativeness: We want a procedure that will not be systematically in error in one direction or another. In technical terms, we want an “unbiased estimate,” if possible. “Unbiased” in this case does not mean “friendly” or “unprejudiced,” but rather implies that on the average — that is, in the long run, after taking repeated samples — estimates that are too high will about balance (in percentage terms) those that are too low. The mean of the universe (or the proportion, if we are speaking of two-valued “binomial situations”) is a frequent object of our interest. And the sample mean is (in most cases) an unbiased estimate of the population mean.</p>
<p>Let’s now see an informal proof that the mean of a randomlydrawn sample is an “unbiased” estimator of the population mean. That is, the errors of the sample means will cancel out after repeated samples because the mean of a large number of sample means approaches the population mean. A second “law” to be informally proven is that the size of the inaccuracy of a sample proportion is largest when the population proportion is near 50 percent, and smallest when it approaches zero percent or 100 percent.</p>
<p>The statement that the sample mean is an unbiased estimate of the population mean holds for <em>many but not all kinds</em> of samples — proportions of two-outcome (Democrat-Republican) events (as in this case) and also the means of many measured-data universes (heights, speeds, and so on) that we will come to later.</p>
<p>But, you object, I have only <em>said</em> that this is so; I haven’t proven it. Quite right. Now we will go beyond this simple assertion, though we won’t reach the level of formal proof. This discussion applies to conventional analytic statistical theory as well as to the resampling approach.</p>
<p>We want to know why the mean of a repeated sample — or the proportion, in the case of a binomial universe — tends to equal the mean of the universe (or the proportion of a binomial sample). Consider a population of one thousand voters. Split the population into random sub-populations of 500 voters each; let’s call these sub-populations by the name “samples.” Almost inevitably, the proportions voting Democratic in the samples will <em>not exactly</em> equal the “true” proportions in the population. (Why not? Well, why should they split evenly? There is no general reason why they should.) But if the sample proportions do <em>not</em> equal the population proportion, we can say that the extent of the difference between the two sample proportions and the population proportion will be <em>identical but in the opposite direction</em>. If the population proportion is 600/1000 = 60 percent, and one sample’s proportion is 340/500 = 68 percent, then the other sample’s proportion must be (600-340 = 260)/500 = 52 percent. So if in the very long run you would choose each of these two samples about half the time (as you would if you selected between the two samples randomly) the average of the sample proportions would be (68 percent + 52 percent)/2 = 60 percent. This shows that on the average the sample proportion is a fair and unbiased estimate of the population proportion — if the sample is half the size of the population.</p>
<p>If we now sub-divide each of our two samples of 500 (each of which was half the population size) into equal-size subsamples of 250 each, the same argument will hold for the proportions of the samples of 250 with respect to the sample of 500: The proportion of a 250-voter sample is an unbiased estimate of the proportion of the 500-voter sample from which it is drawn. It seems inductively reasonable, then, that if the proportion of a 250-voter sample is an unbiased estimate of the 500-voter sample from which it is drawn, and the proportion of a 500-voter sample is an unbiased estimate of the 1000-voter population, then the proportion of a 250-voter sample should be an unbiased estimate of the population proportion. And if so, this argument should hold for samples of 1/2 x 250 = 125, and so on — in fact for <em>any</em> size sample.</p>
<p>The argument given above is not a rigorous formal proof. But I doubt that the non-mathematician needs, or will benefit from, a more formal proof of this proposition. You are more likely to be persuaded if you demonstrate this proposition to yourself experimentally in the following manner:</p>
<ul>
<li><strong>Step 1.</strong> Let “1-6” = Democrat, “7-10” = Republican</li>
<li><strong>Step 2.</strong> Choose a sample of, say, ten random numbers, and record the proportion Democrat (the sample proportion).</li>
<li><strong>Step 3.</strong> Repeat step 2 a thousand times.</li>
<li><strong>Step 4.</strong> Compute the mean of the sample proportions, and compare it to the population proportion of 60 percent. This result should be close enough to reassure you that <em>on the average</em> the sample proportion is an “unbiased” estimate of the population proportion, though in any particular sample it may be substantially off in either direction.</li>
</ul>
</section>
<section id="efficiency" class="level3" data-number="19.3.2">
<h3 data-number="19.3.2" class="anchored" data-anchor-id="efficiency"><span class="header-section-number">19.3.2</span> Efficiency</h3>
<p>We want an estimate to be accurate, in the sense that it is as close to the “actual” value of the parameter as possible. Sometimes it is possible to get more accuracy at the cost of biasing the estimate. More than that does not need to be said here.</p>
</section>
<section id="maximum-likelihood" class="level3" data-number="19.3.3">
<h3 data-number="19.3.3" class="anchored" data-anchor-id="maximum-likelihood"><span class="header-section-number">19.3.3</span> Maximum Likelihood</h3>
<p>Knowing that a particular value is the most likely of all values may be of importance in itself. For example, a person betting on one horse in a horse race is interested in his/her estimate of the winner having the highest possible probability, and is not the slightest bit interested in getting <em>nearly</em> the right horse. Maximum likelihood estimates are of particular interest in such situations.</p>
<p>See <span class="citation" data-cites="savage1972foundations">(<a href="references.html#ref-savage1972foundations" role="doc-biblioref">Savage 1972, chap. 15</a>)</span>, for many other criteria of estimators.</p>
</section>
</section>
<section id="criteria-of-the-criteria" class="level2" data-number="19.4">
<h2 data-number="19.4" class="anchored" data-anchor-id="criteria-of-the-criteria"><span class="header-section-number">19.4</span> Criteria of the Criteria</h2>
<p>What should we look for in choosing criteria? Logically, this question should precede the above list of criteria.</p>
<p>Savage <span class="citation" data-cites="savage1972foundations">(<a href="references.html#ref-savage1972foundations" role="doc-biblioref">1972, chap. 15</a>)</span> has urged that we should always think in terms of the <em>consequences</em> of choosing criteria, in light of our purposes in making the estimate. I believe that he is making an important point. But it often is very hard work to think the matter through all the way to the consequences of the criteria chosen. And in most cases, such fine inquiry is not needed, in the sense that the estimating procedure chosen will be the same no matter what consequences are considered.<a href="#fn1" class="footnote-ref" id="fnref1" role="doc-noteref"><sup>1</sup></a></p>
</section>
<section id="estimation-of-accuracy-of-the-point-estimate" class="level2" data-number="19.5">
<h2 data-number="19.5" class="anchored" data-anchor-id="estimation-of-accuracy-of-the-point-estimate"><span class="header-section-number">19.5</span> Estimation of accuracy of the point estimate</h2>
<p>So far we have discussed how to make a point estimate, and criteria of good estimators. We also are interested in estimating the accuracy of that estimate. That subject — which is harder to grapple with — is discussed in <a href="confidence_1.html" class="quarto-xref"><span>Chapter 26</span></a> and <a href="confidence_2.html" class="quarto-xref"><span>Chapter 27</span></a> on confidence intervals.</p>
<p>Most important: One cannot sensibly talk about the accuracy of probabilities in the abstract, without reference to some set of facts. In the abstract, the notion of accuracy loses any meaning, and invites confusion and argument.</p>
</section>
<section id="sec-uses-of-mean" class="level2" data-number="19.6">
<h2 data-number="19.6" class="anchored" data-anchor-id="sec-uses-of-mean"><span class="header-section-number">19.6</span> Uses of the mean</h2>
<p>Let’s consider when the use of a device such as the mean is valuable, in the context of the data on marksmen in <a href="#tbl-marksmen-scores" class="quarto-xref">Table <span>19.1</span></a>.<a href="#fn2" class="footnote-ref" id="fnref2" role="doc-noteref"><sup>2</sup></a>. If we wish to compare marksman A versus marksman B, we can immediately see that marksman A hit the bullseye (80 shots for 3 points each time) as many times as marksman B hit <em>either</em> the bullseye or simply got in the black (30 shots for 3 points and 50 shots for 2 points), and A hit the black (2 points) as many times as B just got in the white (1 point). From these two comparisons covering all the shots, in both of which comparisons A does better, it is immediately obvious that marksman A is better than marksman B. We can say that A’s score <em>dominates</em> B’s score.</p>
<div id="tbl-marksmen-scores" class="quarto-float quarto-figure quarto-figure-center anchored">
<figure class="quarto-float quarto-float-tbl figure">
<figcaption class="quarto-float-caption-top quarto-float-caption quarto-float-tbl" id="tbl-marksmen-scores-caption-0ceaefa1-69ba-4598-a22c-09a6ac19f8ca">
Table 19.1: Score percentages by marksman
</figcaption>
<div aria-describedby="tbl-marksmen-scores-caption-0ceaefa1-69ba-4598-a22c-09a6ac19f8ca">
<table class="caption-top table">
<colgroup>
<col style="width: 12%">
<col style="width: 23%">
<col style="width: 19%">
</colgroup>
<thead>
<tr class="header">
<th>Score</th>
<th># occurrences</th>
<th>Probability</th>
</tr>
</thead>
<tbody>
<tr class="odd">
<td colspan="3"><strong>Marksman A</strong></td>
</tr>
<tr class="even">
<td>1</td>
<td>0</td>
<td>0</td>
</tr>
<tr class="odd">
<td>2</td>
<td>20</td>
<td>.2</td>
</tr>
<tr class="even">
<td>3</td>
<td>80</td>
<td>.8</td>
</tr>
<tr class="odd">
<td colspan="3"><strong>Marksman B</strong></td>
</tr>
<tr class="even">
<td>1</td>
<td>20</td>
<td>.2</td>
</tr>
<tr class="odd">
<td>2</td>
<td>50</td>
<td>.5</td>
</tr>
<tr class="even">
<td>3</td>
<td>30</td>
<td>.3</td>
</tr>
<tr class="odd">
<td colspan="3"><strong>Marksman C</strong></td>
</tr>
<tr class="even">
<td>1</td>
<td>10</td>
<td>.1</td>
</tr>
<tr class="odd">
<td>2</td>
<td>60</td>
<td>.6</td>
</tr>
<tr class="even">
<td>3</td>
<td>30</td>
<td>.3</td>
</tr>
</tbody>
</table>
</div>
</figure>
</div>
<p>When we turn to comparing marksman C to marksman D, however, we cannot say that one “dominates” the other as we could with the comparison of marksmen A and B. Therefore, we turn to a summarizing device. One such device that is useful here is the mean. For marksman C the mean score is <span class="math inline">\((40 * 1) + (10 * 2) + (50 * 3) =
210\)</span>, while for marksman D the mean score is <span class="math inline">\((10 * 1) + (60 * 2) + (30 * 3) =
220\)</span>. Hence we can say that D is better than C even though D’s score does not dominate C’s score in the bullseye category.</p>
<p>Another use of the mean <span class="citation" data-cites="gnedenko1962elementary">(<a href="references.html#ref-gnedenko1962elementary" role="doc-biblioref">Gnedenko, Aleksandr, and Khinchin 1962, 68</a>)</span> is shown in the estimation of the number of matches that we need to start fires for an operation carried out 20 times in a day (<a href="#tbl-matches-fires" class="quarto-xref">Table <span>19.2</span></a>). Let’s say that the number of cases where s/he needs 1, 2 … 5 matches to start a fire are as follows (along with their probabilities) based on the last 100 fires started:</p>
<div id="tbl-matches-fires" class="quarto-float quarto-figure quarto-figure-center anchored">
<figure class="quarto-float quarto-float-tbl figure">
<figcaption class="quarto-float-caption-top quarto-float-caption quarto-float-tbl" id="tbl-matches-fires-caption-0ceaefa1-69ba-4598-a22c-09a6ac19f8ca">
Table 19.2: Number of matches needed to start a fire
</figcaption>
<div aria-describedby="tbl-matches-fires-caption-0ceaefa1-69ba-4598-a22c-09a6ac19f8ca">
<table class="caption-top table">
<thead>
<tr class="header">
<th>Number of Matches</th>
<th>Number of Cases</th>
<th>Probabilities</th>
</tr>
</thead>
<tbody>
<tr class="odd">
<td>1</td>
<td>7</td>
<td>.16</td>
</tr>
<tr class="even">
<td>2</td>
<td>16</td>
<td>.16</td>
</tr>
<tr class="odd">
<td>3</td>
<td>55</td>
<td>.55</td>
</tr>
<tr class="even">
<td>4</td>
<td>21</td>
<td>.21</td>
</tr>
<tr class="odd">
<td>5</td>
<td>1</td>
<td>.01</td>
</tr>
</tbody>
</table>
</div>
</figure>
</div>
<p>If you know that the operator will be lighting twenty fires, you can estimate the number of matches that s/he will need by multiplying the mean number of matches (which turns out be <span class="math inline">\(1 * .07 + 2 * 0.16 + 3 * 0.55 + 4 * 0.21 + 5 *
0.01 = 2.93\)</span>) in the observed experience by 20. Here you are using the mean as an indication of a representative case.</p>
<p>It is common for writers to immediately produce the data in the forms of percentages or probabilities. But I think it is important to include in our discussion the absolute numbers, because this is what one must begin with in practice. And keeping the absolute numbers in mind is likely to avoid some confusions that arise if one immediately goes to percentages or to probabilities.</p>
<p>Still another use for the mean is when you have a set of observations with error in them. The mean of the observations probably is your best guess about which is the “right” one. Furthermore, the distance you are likely to be off the mark is less if you select the mean of the observations. An example might be a series of witnesses giving the police their guesses about the height of a man who overturned an outhouse. The mean probably is the best estimate to give to police officers as a description of the perpetrator (though it would be helpful to give the range of the observations as well).</p>
<p>We use the mean so often, in so many different circumstances, that we become used to it and never think about its nature. So let’s do so a bit now.</p>
<p>Different statistical ideas are appropriate for business and engineering decisions, biometrics, econometrics, scientific explanation (the philosophers’ case), and other fields. So nothing said here holds everywhere and always.</p>
<p>One might ask: What is the “meaning” of a mean? But that is not a helpful question. Rather, we should ask about the <em>uses</em> of a mean. Usually a mean is used to summarize a set of data. As we saw with marksmen C and D, it often is difficult to look at a table of data and obtain an overall idea of how big or how small the observations are; the mean (or other measurements) can help. Or if you wish to compare two sets of data where the distributions of observations overlap each other, comparing the means of the two distributions can often help you better understand the matter.</p>
<p>Another complication is the confusion between <em>description</em> and <em>estimation</em>, which makes it difficult to decide where to place the topic of descriptive statistics in a textbook. For example, compare the mean income of all men in the U. S., as measured by the decennial census. This mean of the universe can have a very different meaning from the mean of a sample of men with respect to the same characteristic. The sample mean is a point estimate, a statistical device, whereas the mean of the universe is a description. The use of the mean as an estimator is fraught with complications. Still, maybe it is no more complicated than deciding what describer to use for a population. This entire matter is much more complex than it appears at first glance.</p>
<p>When the sample size approaches in size the entire population — when the sample becomes closer and closer to being the <em>same</em> as the population — the two issues blend. What does that tell us? Anything? What is the relationship between a baseball player’s average for two weeks, and his/her lifetime average? This is subtle stuff — rivaling the subtleness of arguments about inference versus probability, and about the nature of confidence limits (see <a href="confidence_1.html" class="quarto-xref"><span>Chapter 26</span></a> and <a href="confidence_2.html" class="quarto-xref"><span>Chapter 27</span></a> ). Maybe the only solid answer is to try to stay super-clear on what you are doing for what purpose, and to ask continually what job you want the statistic (or describer) to do for you.</p>
<p>The issue of the relationship of sample size to population size arises here. If the sample size equals or approaches the population size, the very notion of estimation loses its meaning.</p>
<p>The notion of “best estimator” makes no sense in some situations, including the following: a) You draw one black ball from a bucket. You cannot put confidence intervals around your estimate of the proportion of black balls, except to say that the proportion is somewhere between 1 and 0. No one would proceed without bringing in more information. That is, when there is almost no information, you simply cannot make much of an estimate — and the resampling method breaks down, too. It does not help much to shift the discussion to the models of the buckets, because then the issue is the unknown population of the buckets, in which case we need to bring in our general knowledge. b) When the sample size equals or is close to the population size, as discussed in this section, the data are a description rather than an estimate, because the sample is getting to be much the same as the universe; that is, if there are twelve people in your family, and you randomly take a sample of the amount of sugar used by eight members of the family, the results of the sample cannot be very different than if you compute the amount for all twelve family members. In such a case, the interpretation of the mean becomes complex.</p>
<p>Underlying all estimation is the assumption of continuation, which follows from random sampling — that there is no reason to expect the next sample to be different from the present one in any <em>particular</em> fashion, mean or variation. But we do expect it to be different in <em>some</em> fashion because of sampling variability.</p>
</section>
<section id="conclusion" class="level2" data-number="19.7">
<h2 data-number="19.7" class="anchored" data-anchor-id="conclusion"><span class="header-section-number">19.7</span> Conclusion</h2>
<p>A <em>Newsweek</em> article says, “According to a recent reader’s survey in Bride’s magazine, the average blowout [wedding] will set you back about $16,000” (Feb 15, 1993, p. 67). That use of the mean (I assume) for the average, rather than the median, could cost the parents of some brides a pretty penny. It could be that the cost for the average <em>person</em> — that is, the median expenditure — might be a lot less than $16,000. (A few million dollar weddings could have a huge effect on a survey mean.) An inappropriate standard of comparison might enter into some family discussions as a result of this article, and cause higher outlays than otherwise. This chapter helps one understand the nature of such estimates.</p>
<div id="refs" class="references csl-bib-body hanging-indent" data-entry-spacing="0" role="list" style="display: none">
<div id="ref-gnedenko1962elementary" class="csl-entry" role="listitem">
Gnedenko, Boris Vladimirovich, I Aleksandr, and Akovlevich Khinchin. 1962. <em>An Elementary Introduction to the Theory of Probability</em>. New York, NY, USA: Dover Publications, Inc. <a href="https://archive.org/details/gnedenko-khinchin-an-elementary-introduction-to-the-theory-of-probability">https://archive.org/details/gnedenko-khinchin-an-elementary-introduction-to-the-theory-of-probability</a>.
</div>
<div id="ref-savage1972foundations" class="csl-entry" role="listitem">
Savage, Leonard J. 1972. <em>The Foundations of Statistics</em>. New York: Dover Publications, Inc.
</div>
</div>
</section>
<section id="footnotes" class="footnotes footnotes-end-of-document" role="doc-endnotes">
<hr>
<ol>
<li id="fn1"><p>Here I shall merely mention that the method of moments and the method of maximum likelihood serve most of our needs, and often agree in their conclusions; furthermore, we often know when the former may be inappropriate.<a href="#fnref1" class="footnote-back" role="doc-backlink">↩︎</a></p></li>
<li id="fn2"><p>This discussion follows <span class="citation" data-cites="gnedenko1962elementary">(<a href="references.html#ref-gnedenko1962elementary" role="doc-biblioref">Gnedenko, Aleksandr, and Khinchin 1962, chap. 8</a>)</span>.<a href="#fnref2" class="footnote-back" role="doc-backlink">↩︎</a></p></li>
</ol>
</section>
</main> <!-- /main -->
<script id="quarto-html-after-body" type="application/javascript">
window.document.addEventListener("DOMContentLoaded", function (event) {
const toggleBodyColorMode = (bsSheetEl) => {
const mode = bsSheetEl.getAttribute("data-mode");
const bodyEl = window.document.querySelector("body");
if (mode === "dark") {
bodyEl.classList.add("quarto-dark");
bodyEl.classList.remove("quarto-light");
} else {
bodyEl.classList.add("quarto-light");
bodyEl.classList.remove("quarto-dark");
}
}
const toggleBodyColorPrimary = () => {
const bsSheetEl = window.document.querySelector("link#quarto-bootstrap");
if (bsSheetEl) {
toggleBodyColorMode(bsSheetEl);
}
}
toggleBodyColorPrimary();
const icon = "";
const anchorJS = new window.AnchorJS();
anchorJS.options = {
placement: 'right',
icon: icon
};
anchorJS.add('.anchored');
const isCodeAnnotation = (el) => {
for (const clz of el.classList) {
if (clz.startsWith('code-annotation-')) {
return true;
}
}
return false;
}
const onCopySuccess = function(e) {
// button target
const button = e.trigger;
// don't keep focus
button.blur();
// flash "checked"
button.classList.add('code-copy-button-checked');
var currentTitle = button.getAttribute("title");
button.setAttribute("title", "Copied!");
let tooltip;
if (window.bootstrap) {
button.setAttribute("data-bs-toggle", "tooltip");
button.setAttribute("data-bs-placement", "left");
button.setAttribute("data-bs-title", "Copied!");
tooltip = new bootstrap.Tooltip(button,
{ trigger: "manual",
customClass: "code-copy-button-tooltip",
offset: [0, -8]});
tooltip.show();
}
setTimeout(function() {
if (tooltip) {
tooltip.hide();
button.removeAttribute("data-bs-title");
button.removeAttribute("data-bs-toggle");
button.removeAttribute("data-bs-placement");
}
button.setAttribute("title", currentTitle);
button.classList.remove('code-copy-button-checked');
}, 1000);
// clear code selection
e.clearSelection();
}
const getTextToCopy = function(trigger) {
const codeEl = trigger.previousElementSibling.cloneNode(true);
for (const childEl of codeEl.children) {
if (isCodeAnnotation(childEl)) {
childEl.remove();
}
}
return codeEl.innerText;
}
const clipboard = new window.ClipboardJS('.code-copy-button:not([data-in-quarto-modal])', {
text: getTextToCopy
});
clipboard.on('success', onCopySuccess);
if (window.document.getElementById('quarto-embedded-source-code-modal')) {
// For code content inside modals, clipBoardJS needs to be initialized with a container option
// TODO: Check when it could be a function (https://github.com/zenorocha/clipboard.js/issues/860)
const clipboardModal = new window.ClipboardJS('.code-copy-button[data-in-quarto-modal]', {
text: getTextToCopy,
container: window.document.getElementById('quarto-embedded-source-code-modal')
});
clipboardModal.on('success', onCopySuccess);
}
var localhostRegex = new RegExp(/^(?:http|https):\/\/localhost\:?[0-9]*\//);
var mailtoRegex = new RegExp(/^mailto:/);
var filterRegex = new RegExp('/' + window.location.host + '/');
var isInternal = (href) => {
return filterRegex.test(href) || localhostRegex.test(href) || mailtoRegex.test(href);
}
// Inspect non-navigation links and adorn them if external
var links = window.document.querySelectorAll('a[href]:not(.nav-link):not(.navbar-brand):not(.toc-action):not(.sidebar-link):not(.sidebar-item-toggle):not(.pagination-link):not(.no-external):not([aria-hidden]):not(.dropdown-item):not(.quarto-navigation-tool):not(.about-link)');
for (var i=0; i<links.length; i++) {
const link = links[i];
if (!isInternal(link.href)) {
// undo the damage that might have been done by quarto-nav.js in the case of
// links that we want to consider external
if (link.dataset.originalHref !== undefined) {
link.href = link.dataset.originalHref;
}
}
}
function tippyHover(el, contentFn, onTriggerFn, onUntriggerFn) {
const config = {
allowHTML: true,
maxWidth: 500,
delay: 100,
arrow: false,
appendTo: function(el) {
return el.parentElement;
},
interactive: true,
interactiveBorder: 10,
theme: 'quarto',
placement: 'bottom-start',
};
if (contentFn) {
config.content = contentFn;
}
if (onTriggerFn) {
config.onTrigger = onTriggerFn;
}
if (onUntriggerFn) {
config.onUntrigger = onUntriggerFn;
}
window.tippy(el, config);
}
const noterefs = window.document.querySelectorAll('a[role="doc-noteref"]');
for (var i=0; i<noterefs.length; i++) {
const ref = noterefs[i];
tippyHover(ref, function() {
// use id or data attribute instead here
let href = ref.getAttribute('data-footnote-href') || ref.getAttribute('href');
try { href = new URL(href).hash; } catch {}
const id = href.replace(/^#\/?/, "");
const note = window.document.getElementById(id);
if (note) {
return note.innerHTML;
} else {
return "";
}
});
}
const xrefs = window.document.querySelectorAll('a.quarto-xref');
const processXRef = (id, note) => {
// Strip column container classes
const stripColumnClz = (el) => {
el.classList.remove("page-full", "page-columns");
if (el.children) {
for (const child of el.children) {
stripColumnClz(child);
}
}
}
stripColumnClz(note)
if (id === null || id.startsWith('sec-')) {
// Special case sections, only their first couple elements
const container = document.createElement("div");
if (note.children && note.children.length > 2) {
container.appendChild(note.children[0].cloneNode(true));
for (let i = 1; i < note.children.length; i++) {
const child = note.children[i];
if (child.tagName === "P" && child.innerText === "") {
continue;
} else {
container.appendChild(child.cloneNode(true));
break;
}
}
if (window.Quarto?.typesetMath) {
window.Quarto.typesetMath(container);
}
return container.innerHTML
} else {
if (window.Quarto?.typesetMath) {
window.Quarto.typesetMath(note);
}
return note.innerHTML;
}
} else {
// Remove any anchor links if they are present
const anchorLink = note.querySelector('a.anchorjs-link');
if (anchorLink) {
anchorLink.remove();
}
if (window.Quarto?.typesetMath) {
window.Quarto.typesetMath(note);
}
// TODO in 1.5, we should make sure this works without a callout special case
if (note.classList.contains("callout")) {
return note.outerHTML;
} else {
return note.innerHTML;
}
}
}
for (var i=0; i<xrefs.length; i++) {
const xref = xrefs[i];
tippyHover(xref, undefined, function(instance) {
instance.disable();
let url = xref.getAttribute('href');
let hash = undefined;
if (url.startsWith('#')) {
hash = url;
} else {
try { hash = new URL(url).hash; } catch {}
}
if (hash) {
const id = hash.replace(/^#\/?/, "");
const note = window.document.getElementById(id);
if (note !== null) {
try {
const html = processXRef(id, note.cloneNode(true));
instance.setContent(html);
} finally {
instance.enable();
instance.show();
}
} else {
// See if we can fetch this
fetch(url.split('#')[0])
.then(res => res.text())
.then(html => {
const parser = new DOMParser();
const htmlDoc = parser.parseFromString(html, "text/html");
const note = htmlDoc.getElementById(id);
if (note !== null) {
const html = processXRef(id, note);
instance.setContent(html);
}
}).finally(() => {
instance.enable();
instance.show();
});
}
} else {
// See if we can fetch a full url (with no hash to target)
// This is a special case and we should probably do some content thinning / targeting
fetch(url)
.then(res => res.text())
.then(html => {
const parser = new DOMParser();
const htmlDoc = parser.parseFromString(html, "text/html");
const note = htmlDoc.querySelector('main.content');
if (note !== null) {
// This should only happen for chapter cross references
// (since there is no id in the URL)
// remove the first header
if (note.children.length > 0 && note.children[0].tagName === "HEADER") {
note.children[0].remove();
}
const html = processXRef(null, note);
instance.setContent(html);
}
}).finally(() => {
instance.enable();
instance.show();
});
}
}, function(instance) {
});
}
let selectedAnnoteEl;
const selectorForAnnotation = ( cell, annotation) => {
let cellAttr = 'data-code-cell="' + cell + '"';
let lineAttr = 'data-code-annotation="' + annotation + '"';
const selector = 'span[' + cellAttr + '][' + lineAttr + ']';
return selector;