GEOG281 Lecture Notes - Lecture 10: Random Variable, Per Capita Income, Spatial Analysis

• School: University of Waterloo
• Department: Geography and Environmental Management
• Course: GEOG281
• Professor: Su Yin Tan

Document Summary

Moran"s i range from -1 to +1 (+1 high spaial (clumping together of similar values hot spot in crime map, disease, etc)) (-1 low spaial values) Slide 18: moran"s i - per capita income. Is there a signiicant spaial dependence in data set? (yes there is) high per capita and low per capita is clumped together. Per capita income is not random, inluenced by planning, zoning, economics, where people live, where businesses are, housing, employment etc. 0. 66= +ve case, very close to perfect 1, that means it is a strong relaionship. P= probability value less than 0. 1% you would get this relaionship, there is signiicant spaial dependence (not random) Moran"s i is very close to 0 meaning 0. 012. P= 0. 515, 51. 5% of the ime that we are going to get a patern like this, prety common, there is no signiicant spaial dependence or autocorrelaion (enirely random patern)