1 / 16

Autocorrelation: variable correlated on itself.

Autocorrelation: variable correlated on itself. Observations that are “proximate” will have similar values (positive autocorrelation). “Proximate” can be defined in many ways: Closer together in time (1 dimension). Temporal autocorrelation.

ewa
Download Presentation

Autocorrelation: variable correlated on itself.

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Autocorrelation: variable correlated on itself. Observations that are “proximate” will have similar values (positive autocorrelation). “Proximate” can be defined in many ways: • Closer together in time (1 dimension). Temporal autocorrelation. • Closer together in space (2 dimensions). Spatial autocorrelation.

  2. Degree of autocorrelation can be calculated: • For dependent or independent variables. • For regression residuals.

  3. Autocorrelation of regression residuals creates problem. • Usual view: estimated coefficients are unbiased, but standard errors are biased. • Alternative view: autocorrelated residuals signal presence of omitted variables.  estimated coefficients are biased.

  4. Residuals: Hedonic House Price Model (Blue paid too little; Red paid too much)

  5. Red: Brick; Blue: No Brick

  6. Soccer Scores

  7. Protein per Capita 1997

  8. TB rate 1997

  9. HIV rate 1997

  10. % children in LF 1995-2000

  11. Fertility Rate 1995-2000

  12. Per Capita GDP 1995-2000

  13. 12 Proximity Matrices • Physical Distance • Language Phylogeny • Religion • Huntington Civlization • Colonial/Imperial • Level of Development • Ecology • Trade • Formal Treaty • Allies • Enemies • Event Frequency

  14. The 12 selected languages are on the periphery of the digraph. Links point toward higher taxonomic levels, with all nodes ultimately connected to the node labeled Indo-European. The numbers indicate for selected nodes the maximum path length leading to that node. The taxonomy is from Grimes 2000

  15. Language Macro-Families

  16. GIS: Ecological Regions

More Related