As we know that, Wavelet variance decomposes variance on a scale by scale basis? So, what kind of conclusion can we draw from the wavelet variance analysis? How does it differ from the normal variance?
Picture_Source: Tian, G., Qiao, Z., & Xu, X. (2014). Characteristics of particulate matter (PM10) and its relationship with meteorological factors during 2001–2012 in Beijing. Environmental pollution, 192, 266-274.
Possible interpretation. Intuitively variance gives an idea of the spread around the mean. Then with wavelets we can see how this spread pattern is changing over time .In the graph shown it would mean high spread around day 300 leading to a likely high spread of particulate matter ..a storm or unusual activity?
Cheers
Hi, Ashok
The wavelet decomposition of signal is multi resolution analysis of signal. The normal variance of signal is sum of wavelet variance in different level. So if the variance is more, then indicate about more uncertainty in that level or more unusual activity (as at 300). In other term, there is more information.
https://academic.oup.com/biomet/article-abstract/82/3/619/260725?redirectedFrom=PDF
Karunesh K. Gupta Muhammad Abbas Khan Thank you for the response. Now, I have become able to understand it. Thank you very much.
- Badges
- Science topic
- Similar topics
- Books