Chapter 14

Question 1

Dependent variable

Answer 1

The variable being predicted, Y

Question 2

Independent variable

Answer 2

Variable or variables used to predict the dependent variable, X

Question 3

Simple linear regression

Answer 3

Regression analysis involving one independent variable and one dependent variable in which the relationship is approximated with a straight line

Question 4

Regression model

Answer 4

The equation that describes how Y is related to x and a error term

Question 5

Least square method

Answer 5

Is a procedure for using sample data to find the estimated regression equation

Question 6

Coefficient of determination

Answer 6

Provides a measure of the goodness of fit for the estimated regression equation

Question 7

Ith residual

Answer 7

The difference between the observed value if the dependent variable and the estimated value of the dependent. The residual represents the error using the estimated regression equation to estimate y.

Question 8

SSE

Answer 8

Sum of squares due to error, like the residuals

Question 9

SST

Answer 9

Total sum of squares, SSR + SSE

Question 10

SSR

Answer 10

Sum of squares due to regression

Question 11

Coefficient of determination

Answer 11

R2 —> SSR/SST

Question 12

Correlation coefficient

Answer 12

Rxy —> Sign of b1 multiplicerat med kvadratroten ur R2.
A descriptive measure of the strength of linear association between X and Y

Question 13

Assumptions about the error term

Answer 13

• the error term is a random variable with mean or expected value of 0
• the variance of the error term is the same for all values of X
• the values of the error term i independent
• the error term is normally distributed random variable

Question 14

Confidence interval

Answer 14

Is an interval estimate of the mean value of Y for a given value of x

Question 15

Prediction interval

Answer 15

Is an interval estimate of an individual value of Y for a given value of x

Question 16

Autocorrelation

Answer 16

Also called serial correlation. When a value for Y at time t is related to the value of Y at a previous time period we have autocorrelation.
Positive autocorrelation - positive residual followed by positive residual or negative followed by negative
Negative autocorrelation - positive residual followed by negative residual or negative followed by positive.
The Durban Watson test can detect first order autocorrelation

Question 17

Influential observation

Answer 17

Observetations with high leverage, the leverage is determined by how far the values of the independent variables are from their mean values