Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
起こりうる誤った推論/平均・分散・標準偏差・自由度 / Possible erroneous ...
Search
Kenji Saito
PRO
December 06, 2023
Business
0
140
起こりうる誤った推論/平均・分散・標準偏差・自由度 / Possible erroneous inferences, and mean, variance, standard deviation, degree of freedom
早稲田大学大学院経営管理研究科「企業データ分析」2023 冬の第3-4回で使用したスライドです。
Kenji Saito
PRO
December 06, 2023
Tweet
Share
More Decks by Kenji Saito
See All by Kenji Saito
続・インクルーシブな社会へ / Continuing Towards an Inclusive Society
ks91
PRO
0
17
AGI (人工一般知能) と創る新しく奇妙な社会 / New and Stranger Society built with AGI
ks91
PRO
0
64
回帰分析/大規模言語モデルと統計 / Regression Analysis, Large Language Models and Statistics
ks91
PRO
0
69
多重比較/相関分析 / Multiple Comparison and Correlation Analysis
ks91
PRO
0
65
アカデミーキャンプ 2025冬「考えるのは奴らだ」 / Academy Camp 2025 Winter - Live and Let Think DAY 3
ks91
PRO
0
61
アカデミーキャンプ 2025冬「考えるのは奴らだ」 / Academy Camp 2025 Winter - Live and Let Think DAY 2
ks91
PRO
0
48
アカデミーキャンプ 2025冬「考えるのは奴らだ」 / Academy Camp 2025 Winter - Live and Let Think DAY 1
ks91
PRO
1
74
インクルーシブな社会へ / Toward an Inclusive Society
ks91
PRO
0
22
P 値と有意差/分散分析 / P-value, Significant Difference and Analysis of Variance
ks91
PRO
0
71
Other Decks in Business
See All in Business
VISASQ: ABOUT DEV TEAM
eikohashiba
3
24k
2022~2025年の成長戦略(アップデート)
junkiogawa
0
2.2k
SHIFT ASIA 会社説明資料 V2.1
shiftasiarec
0
250
Space BD - CULTURE BOOK「道」
spacebd
0
1k
anveil_companydeck_2025
anveilpr
0
330
クラスメソッド_営業向け会社紹介資料_202502 / introduction to classmethod for sales
classmethod_jinji
0
360
総合研究院の研究活動と企業等との連携促進|Science Tokyo(東京科学大学)
sciencetokyo
PRO
0
1.7k
IRIAM Culture Deck
iriam
4
100k
How should you respond to feedback from reviews and tests
kitanosirokuma
2
180
株式会社SAFELY 会社紹介 / Company
safely_pr
1
290
Rimo会社紹介資料
rimollc
1
200
merpay-Overview
mercari_inc
7
170k
Featured
See All Featured
Adopting Sorbet at Scale
ufuk
74
9.2k
The Art of Delivering Value - GDevCon NA Keynote
reverentgeek
10
1.3k
Typedesign – Prime Four
hannesfritz
40
2.5k
Git: the NoSQL Database
bkeepers
PRO
427
65k
The Illustrated Children's Guide to Kubernetes
chrisshort
48
49k
Rebuilding a faster, lazier Slack
samanthasiow
80
8.8k
Raft: Consensus for Rubyists
vanstee
137
6.8k
Writing Fast Ruby
sferik
628
61k
VelocityConf: Rendering Performance Case Studies
addyosmani
328
24k
Performance Is Good for Brains [We Love Speed 2024]
tammyeverts
7
640
Building a Modern Day E-commerce SEO Strategy
aleyda
38
7.1k
Scaling GitHub
holman
459
140k
Transcript
generated by Stable Diffusion XL v1.0 2023 3-4 (WBS) 2023
3-4 — 2023-12-07 – p.1/37
https://speakerdeck.com/ks91/collections/corporate-data-analysis-2023-winter 2023 3-4 — 2023-12-07 – p.2/37
( ) 1 11 30 • 2 11 30 (B
A ) • 3 12 7 • 4 12 7 • 5 12 14 6 12 14 t 7 12 21 2 ( ) t 8 12 21 2 ( ) t 9 1 11 P 10 1 11 11 1 18 12 1 18 13 1 25 14 1 25 W-IOI 2023 3-4 — 2023-12-07 – p.3/37
( 20 ) 1 • 2 R • 3 •
4 • 5 6 ( ) 7 (1) 8 (2) 9 R ( ) (1) 10 R ( ) (2) 11 R ( ) (1) 12 R ( ) (2) 13 GPT-4 14 GPT-4 15 ( ) LaTeX Overleaf 8 (12/14 ) / (2 ) OK / 2023 3-4 — 2023-12-07 – p.4/37
(B A ) 1 ( ) 2 (Wilcoxon-Mann-Whitney ) 2023
3-4 — 2023-12-07 – p.5/37
3 1 ( ) 2 ( ) 1 2 4
σ2 σ s2 s df 2023 3-4 — 2023-12-07 – p.6/37
2023 3-4 — 2023-12-07 – p.7/37
1. (1) (2) 2023 12 3 ( ) 23:59 JST
( ) Waseda Moodle (Q & A ) 2023 3-4 — 2023-12-07 – p.8/37
. . . . . . 13(12) 11(10) (12/5( )
) ( ) R ( ) (1) 2023 3-4 — 2023-12-07 – p.9/37
(1) ( ) 2 1 ⇒ 1 2 (2) 95%
⇒ ( 6 ) 11 2 95% (← be ambitious!) 5% (← 2 ) 2023 3-4 — 2023-12-07 – p.10/37
0 2 4 6 8 10 0.0 0.1 0.2 0.3
0.4 0.5 0.6 ᳨ฟᅇᩘ ☜⋡ 2023 3-4 — 2023-12-07 – p.11/37
W GitHub Gist ⇒ RStudio Git GitHub https://github.com/ks91/cda-demo “cda-demo” RStudio
Git Git Dropbox “cda-demo” 2023 3-4 — 2023-12-07 – p.12/37
Git Git ( GPL) GitHub Git ( ) RStudio pull
Git (OS ) Linux : ( OK) macOS : ( OK) Windows : https://gitforwindows.org OK https://github.com/ks91/cda-demo ( ) 2023 3-4 — 2023-12-07 – p.13/37
S ⇒ D R ( ) × * ÷ /
xy x^y ( ) √ x sqrt(x) (function; ) ( 1 , 2 ,. . .) sqrt(9) + sqrt(16) <- ( ) ( ) x <- x + 1 x 1 ( ) 2023 3-4 — 2023-12-07 – p.14/37
R ( ) ⇒ D R " " T (true;
) F (false; ) c( 1 , 2 , . . .) ( ) 10 x x[1:3] 1 3 ( 1:100 1 100 ) a.b . . . . R R Source “ D.R” . . . ra <- sum(df[df$group == " ", "rank"]) ⇒ df group rank ra ( == ) 2023 3-4 — 2023-12-07 – p.15/37
“ D.R” ‘sum(. . .)’ # sum(df[df$group == " ",
"rank"]) ... <- 0 i <- 1 # while (i df ) { # if ((df i group ) == " ") { <- + (df i rank ) } i <- i + 1 } 2023 3-4 — 2023-12-07 – p.16/37
S ( ) ⇒ ↑ ( ) : Amazon PPT
Discord 2023 3-4 — 2023-12-07 – p.17/37
S 2 (A B ) ⇒ ( ) ( )
2023 3-4 — 2023-12-07 – p.18/37
S 1 1 1/6 5/6 1 1/6 ⇒ n 1
x ( n = 10 ) 2023 3-4 — 2023-12-07 – p.19/37
0 2 4 6 8 10 0.0 0.1 0.2 0.3
᳨ฟᅇᩘ ☜⋡ 2023 3-4 — 2023-12-07 – p.20/37
S A 5 B A ( ) A 5 B
A ( ) ( ) ( ) 1% 5 1/3 5 =0.41% 1% ⇒ n = 5 x = 5 x (1 3 )5 2023 3-4 — 2023-12-07 – p.21/37
0 1 2 3 4 5 0.0 0.1 0.2 0.3
᳨ฟᅇᩘ ☜⋡ 5% 2023 3-4 — 2023-12-07 – p.22/37
S 120g (120g ) 120g ⇒ 1kg 950g S 120g
· 120g 120g 2023 3-4 — 2023-12-07 – p.23/37
S ⇒ (p.84) χ2 H0 H0 [ ] RStudio Help
“chi” “chisq.test” 2023 3-4 — 2023-12-07 – p.24/37
χ2 1 6 (H0 ) # ( H_0 ) >
chisq.test(c(28, 33, 25, 35, 27, 32), p=c(1/6, 1/6, 1/6, 1/6, 1/6, 1/6)) # ( H_0 ) > chisq.test(c(42, 33, 25, 35, 27, 18), p=c(1/6, 1/6, 1/6, 1/6, 1/6, 1/6)) ‘chisq.test(. . .)’ ( ) ( ) ( ) ← R χ2 2023 3-4 — 2023-12-07 – p.25/37
K ⇒ 2023 3-4 — 2023-12-07 – p.26/37
. . . 2023 3-4 — 2023-12-07 – p.27/37
3 1 ( ) 2 ( ) 1 2 2023
3-4 — 2023-12-07 – p.28/37
100% 2 1 ( ( ) ) 2 ( (
) ) 2 1 1 α (α ) ( ) 2 β (β ) α n p 1 − β (power) 2023 3-4 — 2023-12-07 – p.29/37
3 “ .R” 3 bidist(n, p) : binull(n, p) :
5% bidistg(n, p0, p) : p0 p ( n = 20) p = 0.6 p = 0.2 p = 0.4 p = 0.8 p = 0.7 2023 3-4 — 2023-12-07 – p.30/37
0 5 10 15 20 0.00 0.05 0.10 0.15 0.20
᳨ฟᅇᩘ ☜⋡ 0 5 10 15 20 0.00 0.05 0.10 0.15 0.20 p = 0.2 2023 3-4 — 2023-12-07 – p.31/37
4 σ2 σ s2 s df 2023 3-4 — 2023-12-07
– p.32/37
(parameter; ) ( ) µ, σ2, σ (statistic) ( )
x, s2, s (degree of freedom) ( ) df = n− k 2023 3-4 — 2023-12-07 – p.33/37
E ( p.106) 8 “ E.R” R ‘var(. . .)’,
‘sd(. . .)’ x sqrt(sd(x)^2*(length(x) - 1)/length(x)) 2023 3-4 — 2023-12-07 – p.34/37
2023 3-4 — 2023-12-07 – p.35/37
2. 1 2 (1) 1 2 (2) 2023 12 10
( ) 23:59 JST ( ) Waseda Moodle (Q & A ) (1) Discord 2023 3-4 — 2023-12-07 – p.36/37
2023 3-4 — 2023-12-07 – p.37/37