Week 1 content Flashcards by Arnzie Haz

When we study the relationship between two variables, what do we refer to?

a bivariate analysis

How well did you know this?

Not at all

Perfectly

What graphical technique shows the relationship between variables?

scatter diagram

How well did you know this?

Not at all

Perfectly

What do you need to draw a scatter diagram?

two variables

scale one variable along the horizontal axis (X axis)

scale the other variable along the vertical axis (Y axis)

How well did you know this?

Not at all

Perfectly

What is the dependent variable?

the variable being predicted or estimated

How well did you know this?

Not at all

Perfectly

What is the independent variable?

the predictor variable

it provides the basis for estimation

How well did you know this?

Not at all

Perfectly

What is the coefficient of correlation also known as?

How well did you know this?

Not at all

Perfectly

What is the coefficient of correlation (r)?

a measure of the strength of the relationship between two variables

it can range from -1 to +1

How well did you know this?

Not at all

Perfectly

When r = -1 or 1, what does this indicate?

perfect and strong correlation

How well did you know this?

Not at all

Perfectly

If r = -1, and there is a negative slope, what is the correlation?

perfect negative correlation

How well did you know this?

Not at all

Perfectly

If r = =1, and there is a positive slope, what is the correlation?

perfect positive correlation

How well did you know this?

Not at all

Perfectly

What do positive values of r indicate?

a direct relationship

eg there are two variables, as values assumed by A increase, then the values of B increase as well

How well did you know this?

Not at all

Perfectly

What do negative values of r indicate?

an inverse relationship

eg there are two variables e and f, as the values assumed by e increase, then the values of the f decrease

How well did you know this?

Not at all

Perfectly

What is the equation for r?

r = ∑(Xi - X̄) (Yi - ȳ) / (n-1) SxSy

How well did you know this?

Not at all

Perfectly

What does the correlation coefficient (r) depend on?

r depends entirely on dispersion

the product of the total dispersions of each variable

the product of the standard deviations of each variable

How well did you know this?

Not at all

Perfectly

What does x̄ = ?

x̄ =(ΣXi) / n

the mean of variable X

How well did you know this?

Not at all

Perfectly

What does Sx = ?

Sx=√((ΣXi - x̄)^2 / √(n-1))

the standard deviation of variable X

How well did you know this?

Not at all

Perfectly

What does ȳ = ?

ȳ=(∑Yi)/n

the mean of variable Y

What does Sy = ?

Sy=√((∑Yi- ȳ)^2 / (n-1))

the standard deviation of variable Y

What does n stand for?

the number of observations

What is r if:

n = 10
x̄ = 22
ȳ = 45
Sx = 9.189
Sy = 14.337

r = ∑(Xi - X̄) (Yi - ȳ) / (n-1) SxSy

–> r = 900 /(10-1) (9.189) (14.337)
–> r = 0.759

When using excel, what is the function to calculate (r) the correlation coefficient?

=CORREL()

What does a correlation of 0.759 mean?

it is positive, therefore there’s a positive relationship between the variables

0.759 is close to +1, thus the correlation is strong

What does the knowledge of the existing casual relationship between two variables imply?

relationship between X and Y is described by a linear function

changes in Y are assumed to be related to changes in X

We can predict the value of a dependent variable based on the value of at least one independent variable

we can explain the impact of changes in an independent variable on the dependent variable

What does correlation NOT mean?

causation

What does casual relationship mean?

one variable is determined by another

What is the linear regression model?

an equation with only two variables plus an error term

What does the error term do?

marks the difference between a deterministic equation and a regression equation

What is the linear regression model equation?

Yi = b0 + b1Xi + ei where: Yi = dependent variable b0 = population y intercept b1 = population slope coefficient Xi = independent variable ei = random error term

What is the random error component in the linear regression equation? Yi = b0 + b1Xi + ei

What is the linear component in the linear regression equation? Yi = b0 + b1Xi + ei

b0 + b1Xi

Why do errors occur?

not every point will be on the regression line, most are scattered around it on the contrary, any prediction based on the regression line will be exactly on the line thus, we can expect an error to occur when comparing the true values to the predicted values

What does the simple linear regression model provide?

an estimate of the observed values

What is the simple linear regression equation / prediction line?

Ŷi = b0 + b1Xi where: Ŷ = estimated/predicted Y value for observation i b0 = estimate of the regression intercept b1 = estimate of the regression slope Xi = value of X for observation i

What is b1 and how do you work it out?

the slope b1 = r (Sy / Sx) where: r = the correlation coefficient between Y and X Sy = the standard deviation of Y Sx= the standard deviation of X ȳ = the average of y x̄ = the average of x

What is b0 and how do you work it out?

the intercept b0 = ȳ - b1x̄ where: r = the correlation coefficient between Y and X Sy = the standard deviation of Y Sx= the standard deviation of X ȳ = the average of y x̄ = the average of x

When is the b0 the estimated mean value of Y?

when the value of X is zero

What is b1 being the estimated change in the mean value of Y a result of?

a one-unit increase in X

What is the interpretation of a positive slope?

an increase in X corresponds an increase in Y ΔY = b1ΔX

What is the interpretation of a negative slope?

an increase in X corresponds a decrease in Y ΔY = -b1ΔX

What is the interpretation of the slope (b1) when it equals zero?

there is no relationship between Y and X

What is the difference between the predicted and observed values equal to?

the error term ei = Yi - Ŷi

What can the linear regression model be used to make and how?

predictions if the intercept (b0) and the slope (b1) of the prediction line are known, then the quantitative relationship between the dependent variable (y) and the independent variable (x) is known --> then we can predict the value of Y given a value of X