Do I know you? Pt. 1 Flashcards
ANOVA
Parametric
test whether the values across multiple groups are different
Pearson’s R
Parametric
examine how 2 things are related
Spearman’s Rho
non-parametric
examine how 2 things are related
Two tailed t test
parametric
compare the values between 2 groups
Mann Whitney U
non-parametric
compare the values between 2 groups
Kruskall Wallis
non-parametric
test whether the values across multiple groups are different
Kernel density (3 facts)
- removes statistical noise from data by smoothing it
- uses gaussian weighting (close = more weight)
- good for showing generalized densities of points
Normal gaussian distribution 3 facts
- follows a bell curve
- uses parametric stats
- defined by mean and standard deviation
null hypothesis
no relationship/association/effect in the population
square of the error
explaining difference between observed and expected values
assess accuracy of model
non-stationarity
looking into local variation in geographically weighted regression
cross-k function
correlogram
shows semivariance and covariance on 1 graph
shows correlation at different lags
cross plot
scatterplot of all semivariance values at different lags (doesn’t show the variance line)
k nearest neighbors
selects the closest n neighbors
good because it counts the same number of features each time
kriging 3 facts
- gives you an estimate of the errors
- models spatial variation
- weights are based on distance between points and their overall spatial arrangement
2 stages of kriging
- estimate the structure of the data along with autocorrelation to pick the best model
- interpolate
deletes all objects in the current work space in R
rm(list=ls())