Financial Econometrics and Empirical Market Microstructure
Outlier Dashboards
Below is a mock dashboard report, which ranks stress themes by surprise (i. e., outlier move as measured in standard deviation). Useful reports will have drill down capability, and the ability to view outlier activity in different dimensions such as theme, country, industry, and asset class (Table 2).
|
Table 2 Stress outlier dashboard [not real data]
|
95 % confidence outliers
|
|
|
July 23, 2013
|
|
Summary: Outlier activity on 4 stress scenarios, 3 asset classes, 2 sovereigns, 5 sectors
|
|
Stress Scenarios (4 outliers)
|
Asset Class (3)
|
Sovereign (2)
|
Sector (5)
|
|
Stress Theme Outlier Ranking
|
Outlier (sd)
|
Return
|
Index level
|
|
1
|
China Hard Landing
|
-3.2
|
-3.5 %
|
|
|
CHINA IR
|
-4.1
|
+26 bp
|
Red to black
|
|
CHINA BANKS
|
-3.4
|
-5.20 %
|
|
|
CHINA INDUSTRIALS
|
-2.9
|
-6.10%
|
|
|
AUD/USD
|
-2.1
|
-2.40 %
|
|
|
2
|
Japan Bubble
|
-2.5
|
-1.60 %
|
|
|
N225
|
-2.1
|
-3.40 %
|
|
|
JGB
|
-3.5
|
+ 12 bp
|
|
|
JPY/USD
|
2.1
|
2.30 %
|
|
|
3
|
BRIC slowdown
|
-2.1
|
-3.20 %
|
|
|
4
|
Commodity Bubble
|
-1.9
|
-2.6 %
|
|
|
This sample report indicates a —3.2 standard deviation in the China Hard Landing Scenario, and the drill down shows exceptional movements in interest rates, banks, industrials, and AUD/USD. Other tabs could be viewed to look at outlier activity in other dimensions (e. g., asset class, country, sector).
Anastasiya Byachkova and Artem Simonov Abstract Econophysics is a relatively new discipline. It is one of the most interesting and promising trends in modeling complex economic systems such as financial …
Original definition of fractal was proposed by Mandelbrot with respect to sets. He defined fractal as a mathematical set with fractal dimension is strictly larger than its topological dimension (Mandelbrot …
Risk management is a core discipline in a rapidly changing world. From finance to ecology, we face unprecedented systemic risks from increasingly coupled global systems. Non-linearities render long term predictions …
