lefty_out

> data.frame(y = c(332928,358323,422244,430035,3168,2397,3099,5508,23802,26346,28546,33933,2286,2370,5667,9468,6024,5370,6126,7971,5979,4893,5007,6381,17985,18771,24081,26532),t=rep(c(1996,2001,2006,2013),7),type=rep(c("car","mb","bus","train","other","bike","walk"),each=4)) -> dat
>
> dat$t96 <- dat$t-1996
> dat$type <- relevel(dat$type, "car")
>
> model <- lm(log(y) ~ t96*type, data=dat)
> summary(model)

Call:
lm(formula = log(y) ~ t96 * type, data = dat)

Residuals:
      Min        1Q    Median        3Q       Max
-0.279909 -0.063599  0.003915  0.076055  0.229129

Coefficients:
               Estimate Std. Error t value Pr(>|t|)
(Intercept)   12.727926   0.130850  97.271  < 2e-16 ***
t96            0.016191   0.012862   1.259 0.228661
typebike      -4.152250   0.185049 -22.439 2.25e-12 ***
typebus       -2.655501   0.185049 -14.350 9.13e-10 ***
typemb        -4.896199   0.185049 -26.459 2.35e-13 ***
typeother     -4.125717   0.185049 -22.295 2.45e-12 ***
typetrain     -5.136153   0.185049 -27.756 1.22e-13 ***
typewalk      -2.950006   0.185049 -15.942 2.27e-10 ***
t96:typebike  -0.010925   0.018189  -0.601 0.557700
t96:typebus    0.004381   0.018189   0.241 0.813167
t96:typemb     0.020331   0.018189   1.118 0.282497
t96:typeother  0.002156   0.018189   0.119 0.907337
t96:typetrain  0.075565   0.018189   4.154 0.000973 ***
t96:typewalk   0.008839   0.018189   0.486 0.634536
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.1617 on 14 degrees of freedom
Multiple R-squared:  0.9947,	Adjusted R-squared:  0.9897
F-statistic: 201.2 on 13 and 14 DF,  p-value: 1.395e-13

> anova(model)
Analysis of Variance Table

Response: log(y)
          Df Sum Sq Mean Sq F value    Pr(>F)
t96        1  1.031  1.0307  39.432 2.024e-05 ***
type       6 66.549 11.0915 424.356 5.214e-15 ***
t96:type   6  0.775  0.1292   4.943  0.006513 **
Residuals 14  0.366  0.0261
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1