library(ff)
library(ffbase)
library(biglm)
data(Affairs, package = "AER")
Affairs$ynaffair[Affairs$affairs > 0] <- 1
Affairs$ynaffair[Affairs$affairs == 0] <- 0

gender <- as.ff(c(Affairs$gender),vmode="integer")
age <- as.ff(c(Affairs$age),vmode="double")
yearsmarried <- as.ff(c(Affairs$yearsmarried),vmode="double")
children <- as.ff(c(Affairs$children),vmode="integer")
religiousness <- as.ff(c(Affairs$religiousness),vmode="integer")
education <- as.ff(c(Affairs$education),vmode="integer")
occupation <- as.ff(c(Affairs$occupation),vmode="integer")
rating <- as.ff(c(Affairs$rating),vmode="integer")
ynaffair <- as.ff(c(Affairs$ynaffair),vmode="integer")

ts <- ffdf(ynaffair,gender,age,yearsmarried,children,religiousness,education,occupation,rating)


full <- bigglm.ffdf(ynaffair ~ gender + age + yearsmarried +
    children + religiousness + education + occupation + rating,
    data=ts,family=binomial(),chunksize=5,sandwich=）
summary(full)
Large data regression model: bigglm(ynaffair ~ gender + age + yearsmarried + children + religiousness +
    education + occupation + rating, data = ts, family = binomial(),
    chunksize = 5)
Sample size =  601
                 Coef    (95%     CI)     SE      p
(Intercept)    0.6993 -1.2040  2.6026 0.9517 0.4624
gender         0.2803 -0.1979  0.7585 0.2391 0.2411
age           -0.0443 -0.0808 -0.0078 0.0182 0.0153
yearsmarried   0.0948  0.0303  0.1592 0.0322 0.0033
children       0.3977 -0.1853  0.9807 0.2915 0.1725
religiousness -0.3247 -0.5042 -0.1452 0.0898 0.0003
education      0.0211 -0.0800  0.1221 0.0505 0.6769
occupation     0.0309 -0.1126  0.1745 0.0718 0.6666
rating        -0.4685 -0.6503 -0.2866 0.0909 0.0000


fit.full <- glm(ynaffair ~ gender + age + yearsmarried +
    children + religiousness + education + occupation + rating,
    data = Affairs, family = binomial())
summary(fit.full)
Call:
glm(formula = ynaffair ~ gender + age + yearsmarried + children +
    religiousness + education + occupation + rating, family = binomial(),
    data = Affairs)

Deviance Residuals:
    Min       1Q   Median       3Q      Max  
-1.5713  -0.7499  -0.5690  -0.2539   2.5191  

Coefficients:
              Estimate Std. Error z value Pr(>|z|)   
(Intercept)    1.37726    0.88776   1.551 0.120807   
gendermale     0.28029    0.23909   1.172 0.241083   
age           -0.04426    0.01825  -2.425 0.015301 *  
yearsmarried   0.09477    0.03221   2.942 0.003262 **
childrenyes    0.39767    0.29151   1.364 0.172508   
religiousness -0.32472    0.08975  -3.618 0.000297 ***
education      0.02105    0.05051   0.417 0.676851   
occupation     0.03092    0.07178   0.431 0.666630   
rating        -0.46845    0.09091  -5.153 2.56e-07 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 675.38  on 600  degrees of freedom
Residual deviance: 609.51  on 592  degrees of freedom
AIC: 627.51

Number of Fisher Scoring iterations: 4

该段程序引用自《R语言实战》第十三章，分别使用用bigglm建模，与glm建模。