Week 9 (Multivariate data) Flashcards
univariate data
the simplest form of analysis
based on one variable
bivariate data
complex analysis
data in which analysis is based on two variables per observation simultaneously
multivariate data
more complex
data in which the analysis is based on three or more variables
what type of analysis looks at inter-relationships among three or more variables
multivariable
what are the pros of multivariate analysis
allows you to identify and quantify complex relationships
what are the cons of multivariate analysis
time consuming, not easily understood
which MVA technique
this is dependent of several factors
what is MVA square 1
simple linear regression
what does simple linear regression predict
makes predictions about the values of 1 variable based on the values of the 2nd variable
what does simple LR estimate
straight-line fit to the data that minimize deviations from the line
in the equation Y = a + bX
y stands for
predicted value of variable Y (DV)
what does a stand for in the equation (y=a+bX)
intercept constant
what does b stand for in the equation (y=a+bX)
regression coefficient (slope of the line)
e =
errors in prediction, most are small because correlation b/t X and Y is not perfect
e^2 =
also called “residuals” or e2 to denote unexplained variance (dispersion, spread