>  library(bootstrap)
>  data(cholost)
>  x <- cholost[ , 1]
>  y <- cholost[ , 2]
>
>  plot(x, y)
>  library(modreg)
>  out <- loess(y ~ x,
+      control = loess.control(surface = "direct"))
>  summary(out)
Call:
loess(formula = y ~ x, control = loess.control(surface = "direct"))

Number of Observations: 164
Equivalent Number of Parameters: 4.05
Residual Standard Error: 21.47
Trace of smoother matrix: 4.4

Control settings:
  normalize:  TRUE
  span      :  0.75
  degree   :  2
  family   :  gaussian
  surface  :  direct
>  curve(predict(out, newdata = data.frame(x = x)),
+     add = TRUE, col = "blue")
>

>  out <- loess(y ~ x)
>  curve(predict(out, newdata = data.frame(x = x)),
+     add = TRUE, col = "green")
> help(lowess)
>
> out <- lowess(x, y)
> names(out)
[1] "x" "y"
> lines(out, col = "red")

>  library(mgcv)
This is mgcv 0.8-9
>  out <- gam(y ~ s(x, bs = "cr"))
>  summary(out)

Family: gaussian
Link function: identity

Formula:
y ~ s(x, bs = "cr")

Parametric coefficients:
           Estimate  std. err.    t ratio    Pr(>|t|)
constant      26.69      1.782      14.97    < 2.22e-16

Approximate significance of smooth terms:
            edf       chi.sq     p-value
s(x)      5.671       167.64     < 2.22e-16

R-sq.(adj) = 0.503   Deviance explained =   52%
GCV score = 466.58   Scale est. = 447.6     n = 164
>  curve(predict(out, newdata = data.frame(x = x)),
+      add = TRUE, col = "green")
>  library(sm)
Library `sm', version 2; Copyright (C) 1997, 2000 A.W.Bowman & A.Azzalini
type help(sm) for summary information
>  h <- hcv(x, y, grid = 16)
>  print(h)
[1] 6.056031
>  out <- sm.regression(x, y, h, display = "none")
>  lines(out$eval.points, out$estimate, col = "purple")
>
>  library(KernSmooth)
KernSmooth 2.22 installed
Copyright M. P. Wand 1997
>  h <- dpill(x, y)
>  out <- locpoly(x, y, bandwidth = h)
>  lines(out, col = "red")
>
>  out <- smooth.spline(x, y)
>  print(out)
Call:
smooth.spline(x = x, y = y)

Smoothing Parameter  spar= 0.7161806  lambda= 0.001110886 (12 iterations)
Equivalent Degrees of Freedom (Df): 7.70965
Penalized Criterion: 27843.06
GCV: 464.3896
>  curve(predict(out, x = x)$y, add = TRUE, col = "orange")
>


> data(housing)
> names(housing)
[1] "Sat"  "Infl" "Type" "Cont" "Freq"
> sapply(housing, class)
$Sat
[1] "ordered" "factor"

$Infl
[1] "factor"

$Type
[1] "factor"

$Cont
[1] "factor"

$Freq
[1] "integer"

> sapply(housing, )
> names(housing$Sat)
NULL
> sapply(housing, levels)
$Sat
[1] "Low"    "Medium" "High"

$Infl
[1] "Low"    "Medium" "High"

$Type
[1] "Tower"     "Apartment" "Atrium"    "Terrace"

$Cont
[1] "Low"  "High"

$Freq
NULL

>  out <- glm(Freq ~ Infl * Type * Cont + Sat, data = housing, family = poisson)

> addterm(out, ~ . + Sat : (Infl + Type + Cont), test = "Chisq")
Single term additions

Model:
Freq ~ Infl * Type * Cont + Sat
         Df Deviance    AIC    LRT   Pr(Chi)
<none>        217.46 610.43
Infl:Sat  4   111.08 512.05 106.37 < 2.2e-16 ***
Type:Sat  6   156.79 561.76  60.67 3.292e-11 ***
Cont:Sat  2   212.33 609.30   5.13   0.07708 .
---
Signif. codes:  0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1