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
R を用いた分析(補講) (2) — 人工データの生成 / Analysis using R ...
Search
Kenji Saito
PRO
November 30, 2024
Technology
76
0
Share
R を用いた分析(補講) (2) — 人工データの生成 / Analysis using R (supplementary) (2) - Generating artificial data
早稲田大学大学院経営管理研究科「企業データ分析」2024 冬のオンデマンド教材 第12回で使用したスライドです。
Kenji Saito
PRO
November 30, 2024
More Decks by Kenji Saito
See All by Kenji Saito
インターネットの技術 / Internet technology
ks91
PRO
0
19
大規模言語モデルの原理と使いこなしの原則 / Principles of Large Language Models and How to Use Them Effectively
ks91
PRO
0
8
ワールドカフェI /チューターを改良する / World Café I and Improving the Tutors
ks91
PRO
0
87
研究って何だっけ / What is research anyway?
ks91
PRO
0
15
ブロックチェーンと分散ファイナンス概論 / Introduction to Blockchain and Decentralized Finance
ks91
PRO
0
10
サイバーフィジカル社会とは何か / What Is a Cyber-Physical Society?
ks91
PRO
0
170
本授業におけるエージェント・ハブの使い方 / How to Use the Agent Hub in This Course
ks91
PRO
0
67
LLM とプロンプトエンジニアリング/チューターを定義する / LLMs and Prompt Engineering, and Defining Tutors
ks91
PRO
0
360
Running Coding Agents (Windows WSL Edition)
ks91
PRO
0
44
Other Decks in Technology
See All in Technology
🀄️ on swiftc
giginet
PRO
0
340
3つのボトルネックを解消し、リリースエンジニアリングを再定義した話
nealle
0
410
Master Dataグループ紹介資料
sansan33
PRO
1
4.6k
Sansan Engineering Unit 紹介資料
sansan33
PRO
1
4.2k
JOAI2026講評会資料(近藤佐介)
element138
1
110
BigQuery × dbtでコスト削減した話
rightcode
0
110
みんなの「データ活用」を支えるストレージ担当から持ち込むAWS活用/コミュニティー設計TIPS 10選~「作れる」より、「続けられる」設計へ~
yoshiki0705
0
110
#jawsugyokohama 100 LT11, "My AWS Journey 2011-2026 - kwntravel"
shinichirokawano
0
190
Hello UUID
mimifuwacc
0
130
CDK Insightsで見る、AIによるCDKコード静的解析(+AI解析)
k_adachi_01
2
120
DevOpsDays Tokyo 2026 軽量な仕様書と新たなDORA AI ケイパビリティで実現する、動くソフトウェアを中心とした開発ライフサイクル / DevOpsDays Tokyo 2026
n11sh1
0
110
New CBs New Challenges
ysuzuki
1
180
Featured
See All Featured
How STYLIGHT went responsive
nonsquared
100
6k
How to build an LLM SEO readiness audit: a practical framework
nmsamuel
1
710
Design in an AI World
tapps
0
190
The World Runs on Bad Software
bkeepers
PRO
72
12k
SEO for Brand Visibility & Recognition
aleyda
0
4.5k
Building Applications with DynamoDB
mza
96
7k
Unlocking the hidden potential of vector embeddings in international SEO
frankvandijk
0
760
Collaborative Software Design: How to facilitate domain modelling decisions
baasie
0
190
Joys of Absence: A Defence of Solitary Play
codingconduct
1
340
Navigating the moral maze — ethical principles for Al-driven product design
skipperchong
2
330
Designing Powerful Visuals for Engaging Learning
tmiket
1
330
Impact Scores and Hybrid Strategies: The future of link building
tamaranovitovic
0
250
Transcript
Boxes and whiskers — generated by Stable Diffusion XL v1.0
2024 12 R ( ) (2) — (WBS) 2024 12 R ( ) (2) — — 2024-11 – p.1/14
https://speakerdeck.com/ks91/collections/corporate-data-analysis-2024-winter 2024 12 R ( ) (2) — — 2024-11
– p.2/14
( 20 ) 1 • 2 R • 3 •
4 • 5 • 6 ( ) • 7 (1) • 8 (2) • 9 R ( ) (1) — Welch • 10 R ( ) (2) — χ2 • 11 R ( ) (1) — • 12 R ( ) (2) — • 13 GPT-4 14 GPT-4 15 ( ) LaTeX Overleaf 8 (12/16 ) / (2 ) OK / 2024 12 R ( ) (2) — — 2024-11 – p.3/14
N(µ, σ2) ρ 2 ( : ˆ y = a
+ b1 x1 + b2 x2 + e ) 2024 12 R ( ) (2) — — 2024-11 – p.4/14
N(µ, σ2) “rnorm()” set.seed(173205) # # N(50, 10^2) 100 x
<- rnorm(n=100, mean=50, sd=10) # x # hist(x) mean(x) sd(x) 2024 12 R ( ) (2) — — 2024-11 – p.5/14
Histogram of x x Frequency 10 20 30 40 50
60 70 80 0 5 10 15 20 25 30 35 mean(x) : 50.06994 sd(x) : 10.30096 2024 12 R ( ) (2) — — 2024-11 – p.6/14
ρ 2 (1/2) MASS “mvrnorm()” “ .R” # r =
0.9 # t = 3.7 # r = 15.2 # t = 7.5 # = -0.5 # <- matrix(c( r^2, * t * r, * r * t, t^2 ), nrow=2) 2024 12 R ( ) (2) — — 2024-11 – p.7/14
“mvrnorm()” = S xx S xy S xy S yy
= S xx rS x S y rS x S y S yy ( r = S xy S x S y ) 2 x, y x, y, z, . . . 2024 12 R ( ) (2) — — 2024-11 – p.8/14
ρ 2 (2/2) MASS “mvrnorm()” “ .R” # set.seed(28284) <-
mvrnorm(n=100, mu=c( r, t), Sigma= ) <- pmin(pmax( [,1], 13.0), 19.9) <- pmin(pmax( [,2], 0.0), 20.0) “ [,1]” “ [,2]” plot 2024 12 R ( ) (2) — — 2024-11 – p.9/14
0 5 10 15 20 13 14 15 16 17
18 ㈇ࡢ┦㛵ࡢ 㐌ᙜࡓࡾࡢㄢእ㐠ື㛫 100m㉮ࡢࢱ࣒ (⛊) r : -0.5932345 ( ) -0.5884094 ( ) 2024 12 R ( ) (2) — — 2024-11 – p.10/14
(1/2) “ .R” n <- 50 # a <- 49.4
# ( (158cm ) ) # r_father <- 0.306 mean_father <- 168.78 sd_father <- 3.2 # r_mother <- 0.37 mean_mother <- 155.32 sd_mother <- 2.45 2024 12 R ( ) (2) — — 2024-11 – p.11/14
(2/2) “ .R” <- round(rnorm(n=n, mean=mean_father, sd=sd_father), digits=1) <- round(rnorm(n=n,
mean=mean_mother, sd=sd_mother), digits=1) e <- rnorm(n=n, mean=0, sd=2.8) # <- round(a + r_father * + r_mother * + e, digits=1) 1 “round()” plot 2024 12 R ( ) (2) — — 2024-11 – p.12/14
ፉ㌟㛗 160 165 170 175 152 156 160 164 160
165 170 175 ∗㌟㛗 152 156 160 164 150 154 158 150 154 158 ẕ㌟㛗 : 34.2484 : 0.3545 : 0.4137 : 0.2831 2024 12 R ( ) (2) — — 2024-11 – p.13/14
2024 12 R ( ) (2) — — 2024-11 –
p.14/14