ch15regressionline

# ch15regressionline - Chapter 15: Regression Line

This preview shows pages 1–7. Sign up to view the full content.

Chapter 15: Regression Line Earlier, each problem considered just one variable, say  x =weight, in kg.  Real-life problems, however, often involve several  variables.  e.g. a patient’s condition,  may be dependent on his  temperature,  x , his pulse,  u , and his blood pressure,  v . How to discover the relationship?  Let us just look at the simplest case of two variables,  x   and  y We want to know how  y  depends on  x .

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

View Full Document
Example 1 : Let x = baldness, in % and y = assets, in \$m. Suppose four men yield the following paired data: (a) Summarize the data into a linear model . (b) A 5 th  man is observed to be 60% bald.  Estimate his assets. ) , ( ), , ( ), , ( ), , ( 4 4 3 3 2 2 1 1 y x y x y x y x
Solution : First, draw a scatter diagram to examine initially the relationship between x and y.

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

View Full Document
Secondly, seeing that the data lie roughly along a straight line, we try using a linear model: where “a” is the “y-intercept” and “b” is the “gradient”. Note that while     is the actual asset of the  i- th person,             is the theoretical asset of the  i- th person when     is plugged  into equation (1). Thirdly, we propose to find the most suitable  and  b  so that  the line (1)                is closest to the data.  This is going to be the “line of best fit.” bx a y + = ˆ ) 1 ..( .......... ˆ bx a y + = i i bx a y + = ˆ i x i y
To do this, we first set up n linear equations: We have more equations than the number of unknowns  two ( a  and  b ).  Multiply each equation by 1 ( the coefficient of  a ), and  add; Then multiply each equation by (the coefficient of  b ), and  add; ) 2 ....( .......... ) ( b x na y i i + = ) 3 ....( .......... ) ( ) ( 2 b x a x y x i i i i + =

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

View Full Document
(2) and (3) are the two normal equations. To solve
This is the end of the preview. Sign up to access the rest of the document.

## ch15regressionline - Chapter 15: Regression Line

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online