Linear Regression 2 Flashcards
What tells us how good our linear regression model is?
The error term included in the linear regression equation (+ e).
Of what is the error term included in the regression equation a measure?
Of how spread out the data are around the line of best fit.
How do we find out the numerical values of m and c?
By using the method of least squares.
What is the first step of the linear regression method?
To ascertain the numerical values of m and c.
What is the second step of the linear regression method?
To use the formula Y = mX + c for each of our data points (e.g. X1, X2, etc.) to calculate a set of predicted values (e.g. Y1, Y2, etc.).
What is the third step of the linear regression method?
To calculate the SSR (the sum of squares of residuals which measures the difference between actual data and the model’s predictions).
What is the equation for SSR?
SSR = ∑(Y-Ŷ)2
What is the fourth step of the linear regression method?
To calculate the SST value (the total of the differences between each Y value and the mean Y value).
What is the equation for SST?
SST = ∑(Y-Ȳ)2
In which scenario would SSR be much smaller than SST?
If each actual Y value is much closer to our angled line rather than our flat line.
What indicates that we have a good model (i.e. knowing the value of an independent variable helps us to predict the value of a dependent variable)?
If SSR is much smaller than SST.
The subtraction of which value from SST equals SSM?
SSR
If SSR is much smaller than SST, why would SSM be large?
Because SSM = SST-SSR
What is the ‘goodness of fit’ equation for the linear regression model?
R2 = SSM/ SST
What is the first step involved in carrying out a regression analysis in SPSS?
To select ‘Analyse’- ‘Regression’- ‘Linear’ to fit a straight line.