The Financial Barometer

Every week, I update on this page the global financial turbulence score, which is a multivariate measure of the fluctuation in the benchmark indices of 12 major stock exchanges. This statistics turns out to be correlated with the VIX, the implied volatility index published by the Chicago Board Options Exchange. Its formula and its constituents are described further down on this page. This statistic, which is also known as Mahalanobis distance, was popularized in Finance by Mark Kritzman and Li Yuanzhen (2010).

The table below shows the global financial turbulence scores computed on each of the dates indicated. The global financial turbulence score fell by 31.6 % between May 7 and May 15.

Table: The Latest Global Financial Turbulence Scores.

Date	Score
2020-02-14	1.63
2020-02-20	1.82
2020-02-28	5.22
2020-03-06	3.97
2020-03-13	13.85
2020-03-19	9.34
2020-03-27	9.58
2020-04-03	5.57
2020-04-09	6.17
2020-04-17	4.66
2020-04-24	5.45
2020-04-28	3.71
2020-05-07	7.98
2020-05-15	2.82

The figure below plots the global financial turbulence score and the VIX. The correlation between these two variables is .67. The current level of the turbulence score is above 4.3, which is the level expected during a highly turbulence period.

Figure : Global Financial Turbulence Scores and VIX, Jan 8, 2000 - May 15, 2020.

Dataset and Code

---

The Constituent Indices of the Global Financial Turbulence Score.

Ticker	Description
NYA	New York Stock Exchange composite
IXIC	NASDAQ composite
N225	Nikkei 225 (Japan)
FTAS	UK FTSE All Share
HSI	Hang Seng index (Hong Kong)
N150	Next 150 (European Union)
GSPTSE	S&P/TSX composite index (Canada)
BSESN	S&P BSE SENSEX (India)
GDAXI	DAX performance index (Germany)
AXJO	S&P/ASX 200 (Australia)
SSMI	Swiss exchange mid-cap index
BVSP	IBOVESPA (Brazil)

Formula

$$d_t² = (r_t - μ ) Σ ^-1 (r_t - μ )',

where $d$ denotes the turbulence score, the vector $r$_t lists the current growth rates of the benchmark indices, the vector $$ μ their historical averages, and $$ Σ designates their variance-covariance matrix.