You are working with the text-only light edition of "H.Lohninger: Teach/Me Data Analysis, Springer-Verlag, Berlin-New York-Tokyo, 1999. ISBN 3-540-14743-8". Click here for further information.
| You are working with the text-only light edition of "H.Lohninger: Teach/Me Data Analysis, Springer-Verlag, Berlin-New York-Tokyo, 1999. ISBN 3-540-14743-8". Click here for further information.
|
Table of Contents  Appendix Exercises Mineral Waters |
|
| See also: survey of exercises, missing values, multiple linear regression |   |
As in many other countries, different brands of Austrian mineral water are labeled, including an excerpt of the chemical analysis of the water. The amount of the most important ions, such as sodium, potassion, calcium, magnesium, chloride, sulphate, and others, is printed on the labels. Suppose you have collected several of those labels, with one specific mineral water having no value for sodium on it. Try to estimate the content of sodium of this brand by establishing a (linear) multivariate model. The idea behind this approach is that some ions move parallel to some others (i.e. sodium will correlate to potassium, or hydrogencarbonate).
Use the data set MINWATER
to estimate the content of sodium of the brand "Vöslauer". You may
now go directly to the 
in order to experiment with the data.
Here is the available data for the brand "Vöslauer":
Ca 113.9
K 2.36
Mg 38.9
Cl 21.4
SO4 234.1
HCO3 266.6
solid residues 721.6
Last Update: 2006-Jän-18