PP: Reconstructing time series into a complex network to assess the evolution dynamics of the correlations among energy prices

Purpose

detect the dynamics in time series of their correlation

Methodology

1. calculate correlation coefficients

2. Network construction: 

i. node: AAB, ABC, etc. totally 5^3=125个nodes. 但是不是所有的mode情况都会出现,于是the number of nodes 是个随机数, 并且满足<125.

ii. edge: 如果出现了从mode1到mode2的转变则增加一条mode1---->mode2 的边. 

iii. weight: mode1---->mode2的转变出现的次数. 

3. clustering: 

Results:

1. major modes: high outdegree nodes. 

interpretation: The higher the weighted outdegree, the more important the node is. 

2. evolution dynamics of correlation modes. 

i. total 6 clusters and shows the evolution relations between clusters.

ii. the evolution of one cluster over time. 

........More.......

Lacking: lack the interpretation of each clusters, just showed there are 6 clusters and the transfer probability between clusters. 

Supplementary knowledge:

1. sensitivity analysis; 敏感度分析

sensitivity analysis is the study of how the uncertainty in the output of a mathematical model or system (numerical or otherwise) can be divided and allocated to different sources of uncertainty in its inputs.^[1][2]