Contents

1
2 企 蟆曙一 襾豪?
3


覦覯襷 覲企.

--一危磯, 覈
diabetes <- read.csv("c:\\data\\diabetes.csv", header=T)
str(diabetes)
diabetes.glm <- glm(deaths ~ age + gender, offset=l_popn, family="poisson", data=diabetes)
summary(diabetes.glm)
anova(diabetes.glm)

1 #

  • 蟆郁骸 覦 襷れ 蟆曙( 5% 覩碁) - 覿覿 襭 蟆 蟲襷れ ?
  • (豺伎危 一危)螳 譬覲 蟆曙
  • poisson 螳( 螻, るジ讓曙朱 蠍 蠎襴襯 螳讌 覿, 螳 event 襴, 螳 讌 襯 殊 蟆 焔)

2 企 蟆曙一 襾豪? #

  • event 覦 襯
  • 覈 -> log(u) = ax + by + c
  • 觜 覈 -> log(u/N) = ax + by + c

3 #

[http](http://freesearch.pe.kr/archives/2555) 一危磯ゼ 伎 覲伎.
tmp <- textConnection( 
  "蟲′譴  ′一ろ  
譟	螻狩′	51
螻譟	螻狩′	22
譴譟	螻狩′	43
譟	′	92
螻譟	′	21
譴譟	′	28
譟	觜′	68
螻譟	觜′	9
譴譟	觜′	22") 
x <- read.table(tmp, header=TRUE)
close.connection(tmp)
head(x)

豺伎伎螻 覿
t <- xtabs(~蟲′譴+′一ろ, data=x)
summary(t)

豺伎伎螻 覿 蟆郁骸

> summary(t)
Call: xtabs(formula =  ~ 蟲′譴 + ′一ろ, data = x)
Number of cases in table: 356 
Number of factors: 2 
Test for independence of all factors:
	Chisq = 18.51, df = 4, p-value = 0.0009808
> 
蟆郁骸襯 覲企 蟲′譴 ′一 レ 殊 蟆 覲 . 願 possion regression 企慨.

fit <- glm(~蟲′譴+′一ろ, data=x, family=poisson)
summary(fit)

> summary(fit)

Call:
glm(formula =  ~ 蟲′譴 + ′一ろ, family = poisson, 
    data = x)

Deviance Residuals: 
       1         2         3         4         5         6  
-2.24478   1.17378   2.16834   0.90724   0.08884  -1.52052  
       7         8         9  
 1.18683  -1.54454  -0.77967  

Coefficients:
               Estimate Std. Error z value Pr(>|z|)    
(Intercept)      2.8299     0.1582  17.883  < 2e-16 ***
蟲′譴譟     1.4006     0.1548   9.047  < 2e-16 ***
蟲′譴譴譟     0.5814     0.1732   3.357 0.000787 ***
′一ろ觜′  -0.1585     0.1368  -1.158 0.246791    
′一ろ′     0.1952     0.1254   1.557 0.119474    
---
Signif. codes:  0 *** 0.001 ** 0.01 * 0.05 . 0.1   1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 137.880  on 8  degrees of freedom
Residual deviance:  18.663  on 4  degrees of freedom
AIC: 76.454

Number of Fisher Scoring iterations: 4

>
覘.. 蟆郁骸螳 ′ ̄讌 . 蟆郁骸襯 襾轟企慨. 譟, 螻狩′壱 覈覈 手?

predict(fit, data.frame(蟲′譴=c("譟"), ′一ろ=c("螻狩′")), type="response")
# exp(predict(fit, data.frame(蟲′譴=c("譟"), ′一ろ=c("螻狩′"))))

> predict(fit, data.frame(蟲′譴=c("譟"), ′一ろ=c("螻狩′")), type="response")
       1 
68.75281
69覈 . R襷螻 るジ一 襾轟朱 れ螻 螳 summary() 伎 襾轟朱 .

               Estimate Std. Error z value Pr(>|z|)    
(Intercept)      2.8299     0.1582  17.883  < 2e-16 ***
蟲′譴譟     1.4006     0.1548   9.047  < 2e-16 ***
蟲′譴譴譟     0.5814     0.1732   3.357 0.000787 ***
′一ろ觜′  -0.1585     0.1368  -1.158 0.246791    ---------> 蠍 覲企 覈語 譬る 蟆   .
′一ろ′     0.1952     0.1254   1.557 0.119474    ---------> 蠍 覲企 覈語 譬る 蟆   .
---
Signif. codes:  0 *** 0.001 ** 0.01 * 0.05 . 0.1   1
  • (Intercept) 2.8299
  • 蟲′譴譟 1.4006
  • ′一ろ螻狩′ -> X .
  • 蟆郁骸朱 exp(2.8299 + 1.4006) 覃 68.7516 企蟆 .

...69 51 譬 谿願 . poisson regression 麹讌 蠏碁危.