topic5b

# topic5b - Statistics 512 Applied Linear Models Topic 5b...

This preview shows pages 1–3. Sign up to view the full content.

Statistics 512: Applied Linear Models Topic 5b Topic Overview This topic will cover A case study in multiple regression A variety of approaches to dealing with multiple regression data. Case Study (Green Bullets) Most bullets contain signiﬁcant amounts of lead. In places where many spent bullets (and casings) fall to the ground (e.g., at shooting ranges on military bases), this lead tends to seap into the water supply. Army tried tungsten bullets. Tungsten is not as reactive. However, tungsten still showed up in the water supply. Want to understand which factors aﬀect absorbtion of tungsten into soil. Three measurements We will concentrate on Linear Kd . Data Aggregate results from n = 18 sites (bases) Response ( y ) – Kd tungsten Predictors ( x 1 , ... , x 7 ) – pH, ORP, % ﬁnes, CEC, TOC %, % gravel, % sand Analysis of Predictors pH Frequency 56789 012345 ORP 200 250 300 % Fines 0 2 04 06 08 01 0 0 0123456 CEC 0 5 10 15 20 25 30 35 01234567 1

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
TOC Frequency 0123456 02468 % Gravel 024681 0 024681012 % Sand 0 2 04 06 08 01 0 0 012345 Nothing normal (except possibly pH and ORP ), but nothing too strange. Analysis Between Predictors Lin Kd pH ORP % Fines CEC TOC % Gravel % Sand Obvious relationship between % sand and % fines ( r = - 0 . 9965). Nothing else large (second largest correlation is - 0 . 5678 between pH and ).
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 03/15/2012 for the course STAT 512 taught by Professor Staff during the Fall '08 term at Purdue.

### Page1 / 8

topic5b - Statistics 512 Applied Linear Models Topic 5b...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online