1 / 60

Environmental Data Analysis with MatLab

Environmental Data Analysis with MatLab. Lecture 17: Covariance and Autocorrelation. SYLLABUS.

hestia
Download Presentation

Environmental Data Analysis with MatLab

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. Environmental Data Analysis with MatLab Lecture 17: • Covariance and Autocorrelation

  2. SYLLABUS Lecture 01 Using MatLabLecture 02 Looking At DataLecture 03Probability and Measurement ErrorLecture 04 Multivariate DistributionsLecture 05Linear ModelsLecture 06 The Principle of Least SquaresLecture 07 Prior InformationLecture 08 Solving Generalized Least Squares ProblemsLecture 09 Fourier SeriesLecture 10 Complex Fourier SeriesLecture 11 Lessons Learned from the Fourier Transform Lecture 12 Power Spectral DensityLecture 13 Filter Theory Lecture 14 Applications of Filters Lecture 15 Factor Analysis Lecture 16 Orthogonal functions Lecture 17 Covariance and AutocorrelationLecture 18 Cross-correlationLecture 19 Smoothing, Correlation and SpectraLecture 20 Coherence; Tapering and Spectral Analysis Lecture 21 InterpolationLecture 22 Hypothesis testing Lecture 23 Hypothesis Testing continued; F-TestsLecture 24 Confidence Limits of Spectra, Bootstraps

  3. purpose of the lecture apply the idea of covariance to time series

  4. Part 1correlations between random variables

  5. Atlantic Rock Dataset Scatter plot of TiO2 and Na2O Na2O TiO2

  6. idealization as a p.d.f. scatter plot Na2O d1 TiO2 d2

  7. types of correlations positive correlation negative correlation uncorrelated d2 d2 d2 d1 d1 d1

  8. the covariance matrix … σ1,2 σ1,3 σ12 … σ22 σ1,2 σ2,3 C = σ2,3 σ32 … σ1,3 … … … …

  9. recall the covariance matrix … σ1,2 σ1,3 σ12 … σ22 σ1,2 σ2,3 C = σ2,3 σ32 … σ1,3 … … … … covariance of d1 and d2

  10. estimating covariance from data

  11. divide (di, dj) plane into bins bin, s dj Ddj di Ddi

  12. estimate total probability in bin from the data: Ps = p(di, dj) ΔdiΔdj≈ Ns/N bin, s with Ns data dj Ddj di Ddi

  13. approximate integral with sum and use p(di, dj) ΔdiΔdj≈Ns/N

  14. approximate integral with sum and use p(di, dj) ΔdiΔdj≈Ns/N shrink bins so no more than one data point in each bin

  15. “sample” covariance

  16. normalize to range ±1 “sample” correlation coefficient

  17. Rij-1perfect negative correlation0no correlation1perfect positive correlation

  18. |Rij| of the Atlantic Rock Dataset SiO2 TiO2 Al2O2 FeO MgO CaO Na2O K2O SiO2 TiO2 Al2O2 FeO MgO CaO Na2O K2O

  19. |Rij| of the Atlantic Rock Dataset 0.73 SiO2 TiO2 Al2O2 FeO MgO CaO Na2O K2O SiO2 TiO2 Al2O2 FeO MgO CaO Na2O K2O

  20. Atlantic Rock Dataset Scatter plot of TiO2 and Na2O Na2O TiO2

  21. Part 2correlations between samples within a time series

  22. Neuse River Hydrograph A) time series, d(t) d(t), cfs time t, days

  23. high degree of short-term correlationwhat ever the river was doing yesterday, its probably doing today, toobecause water takes time to drain away

  24. Neuse River Hydrograph A) time series, d(t) d(t), cfs time t, days

  25. low degree of intermediate-term correlationwhat ever the river was doing last month, today it could be doing something completely differentbecause storms are so unpredictable

  26. Neuse River Hydrograph A) time series, d(t) d(t), cfs time t, days

  27. moderate degree of long-term correlationwhat ever the river was doing this time last year, its probably doing today, toobecause seasons repeat

  28. Neuse River Hydrograph A) time series, d(t) d(t), cfs time t, days

  29. 1 day 3 days 30 days

  30. Let’s assume different samples in time series are random variables and calculate their covariance

  31. usual formula for the covariance assuming time series has zero mean

  32. now assume that the time series is stationary(statistical properties don’t vary with time)so that covariance depends only ontime lag between samples

  33. time series of length Ntime lag of (k-1)Δt

  34. using the same approximation for the sample covariance as before 1

  35. using the same approximation for the sample covariance as before 1 autocorrelation at lag (k-1)Δt

  36. autocorrelation in MatLab

  37. Autocorrelation on Neuse River Hydrograph

  38. Autocorrelation on Neuse River Hydrograph symmetric about zero a point in a time series correlates equally well with another in the future and another in the past

  39. Autocorrelation on Neuse River Hydrograph peak at zero lag a point in time series is perfectly correlated with itself

  40. Autocorrelation on Neuse River Hydrograph falls off rapidly in the first few days lags of a few days are highly correlated because the river drains the land over the course of a few days

  41. Autocorrelation on Neuse River Hydrograph negative correlation at lag of 182 days points separated by a half year are negatively correlated

  42. Autocorrelation on Neuse River Hydrograph positive correlation at lag of 360 days points separated by a year are positively correlated

  43. Autocorrelation on Neuse River Hydrograph repeating pattern A) B) the pattern of rainfall approximately repeats annually

  44. autocorrelation similar to convolution

  45. autocorrelation similar to convolution note difference in sign

  46. Important Relation #1autocorrelation is the convolution of a time series with its time-reversed self

  47. integral form of autocorrelation

  48. integral form of autocorrelation change of variables, t’=-t

More Related