Software to Simplify Statistical Practice ...

Local Polynomial Regression Fitting

 

Data

cars: Speed and Stopping Distances of Cars
Source: Stats Package

 

Analysis

  1. Open the DataBook Cars.vstz
  2. Choose the menu Analyze and the command Local
  3. In Dependent, select dist.
  4. In Independent, select speed
  5. Click OK

 

 

Output

By default, R code is printed after parsing and before evaluation.
You can avoid this in File / Option / Advanced menu.

DataSheet is converted to DataFrame, in the form DataBook.DataSheet

 

> LOESSModel <- loess(formula = dist ~ speed, data = Cars.Sheet1, family = 'gaussian', degree = , normalize = T, control = loess.control(surface = 'interpolate', statistics = 'approximate', tr .... [TRUNCATED]

 

> summary(LOESSModel)
Call:
loess(formula = dist ~ speed, data = Cars.Sheet1, na.action = na.exclude,
   
span = 0.75, normalize = T, family = "gaussian", control = loess.control(surface = "interpolate",
       
statistics = "approximate", trace.hat = "exact", cell = 0.2,
       
iterations = 4))

 

Number of Observations: 50
Equivalent Number of Parameters: 4.78
Residual Standard Error: 15.29
Trace of smoother matrix: 5.24

 

Control settings:
 
normalize:  TRUE
 
span     :  0.75
 
degree   :  2
 
family   :  gaussian
 
surface  :  interpolate    cell = 0.2

 

> summary(residuals(LOESSModel))

    Min.  1st Qu.   Median     Mean  3rd Qu.     Max.
-24.4600  -8.5130  -3.2170  -0.2912   6.6350  42.8300