1
Time Series Analysis in R
2
Introduction to Timeseries
2.1
Background
2.2
Data Wrangling
2.2.1
Load packages
2.2.2
Load data
2.2.3
Format data
2.2.4
First viz
2.3
Trend Extraction
2.3.1
Linear trend model
2.3.2
Quadratic fit
2.3.3
Harmonic waves
2.3.4
Automated model
2.4
Forecasting
3
Signal Processing in R
3.1
Data and packages
3.1.1
Packages
3.1.2
Data
3.2
Filtering
3.2.1
Data spacing
3.2.2
Lowpass filter
3.2.3
High-pass filter
3.2.4
Band-pass filter
3.2.5
Notch filter
3.3
Detrending
3.3.1
Detrending methods in R
3.3.2
SSA detrending
3.4
Comparing methods
3.5
Takeways
4
Association
4.1
Measures of association: NAO vs SOI
4.1.1
Load data and packages
4.1.2
Format and plot the NAO data
4.1.3
Merge the SOI data into the NAO data.frame
4.1.4
Interpolation
4.1.5
Correlation
4.2
Spurious Correlations
4.2.1
Smoothing
4.3
Takeways
4.4
References
5
Spectral Analysis in R
5.1
Load Data and Packages
5.1.1
Packages
5.1.2
Dataset
5.2
Data preview
5.3
Data cleaning
5.3.1
Aggregate to monthly
5.3.2
Even sampling
5.3.3
Spectral analysis
5.3.4
STL
5.3.5
Estimation of scaling behavior
5.4
Gap-tolerant spectral analysis
5.4.1
Lomb-Scargle
5.4.2
Wavelet
5.5
Takeaways
6
Wavelet Transform in R
6.1
Data and packages
6.1.1
Packages
6.1.2
Data description
6.2
Loading and visualizing the
\(\delta\)
D series
6.3
Sampling interval
6.4
Wavelet Analysis
6.4.1
Null hypothesis
6.5
Temperature vs CO2
6.6
Bringing the datasets together
6.6.1
Standardization
6.6.2
Coherency
6.6.3
Common power
6.7
Takeways
3.5
Takeways
R provides powerful packages for signal processing, including filtering and detrending.
Interactive visualization is a key part of the process
, to make sure that the signal processing is achieving what it is supposed to do.