Regression Flashcards
Coefficient of determination
r squared
how much of the variance in y is described by the variance in x
how well can we predict y
Akaike information criterion (AIC)
the lower the value the better
used to compare 2 models to see which is a better fit
Exponential model
calculates a value for y based on Euler’s number (2.7182
y = ae^bx
Fixed vs. adaptive weighting
Fixed = uses the same distance
adaptive = relies on number of nearest neighbors for weighting in GWR
Geographically weighted regression
regression + change in regression through space
relationship between x and y can change across space
estimates local intercept/slope or coefficients for each location
intercept
where regression line hits x = 0
predicted value of y when x = 0
linear regression
can calculate the best fit line for 2 varialbes
y= mx + b (m = slope, b = intercept)
what is the m in y=mx+b?
for every unit change in x there is a ___ change in y
Logarithmic model
fitting a model that includes a logarithmic transformation
Non-linear regression
when a normal regression line does not fit the data or explain it
have to fit it to a different transformation
Polynomial (quadratic model)
fitting a model that involves a quadratic term
predicted values
estimates or predicted dependent var values based on the values of the independent variables
slope
slope of the regression line
for every unit increase in x there is a ___ increase/decrease in y
