Data Matters - HSBC.PDF

2 May 2019

Data Matters

Multi-Asset Quant

Risk-on, Goldilocks-off



The current environment looks like a classic risk-on risk-on move…



…rather than a Goldilocks scenario



April saw a pattern of positive returns from risk-on assets and and negative returns from risk-off assets

The consensus is increasingly looking for a period of stable global growth but low inflation. This, coupled with the sharp rebound in all asset classes at the beginning of 2019, has cemented the belief of many market participants in a so-called Goldilocks scenario. However, when we look at the pattern of asset returns this year, it appears that the true Goldilocks period was confined to January. In contrast, the returns seen over April were more consistent consistent with a classic risk-on risk-on move with positive positive returns from riskriskon asset classes such as equities, and negative returns from typical safe havens such as Developed Market rates.

Correlation This risk-on pattern of returns has been accompanied by an increase in the HSBC RORO Indicator, which measures the strength of cross-asset correlations. If we drill

Mark McDonald Head of Data Science and Analytics HSBC Bank plc [email protected] +44 20 7991 5966 Max Kettner Multi-Asset Strategist HSBC Bank plc [email protected] +44 20 7991 5045 Subhrajit Banerjee, CFA Fixed Income Strategist HSBC Bank plc [email protected] +44 20 7991 6851 Dominic Kini Analyst, Quantitative Credit Strategy HSBC Bank plc [email protected] +44 20 7991 5599 Alastair Pinder, CFA Equity Strategist HSBC Securities (USA) Inc. [email protected] +1 212 525 4131 Peter Barnshaw Analyst, Credit Strategy HSBC Bank plc [email protected] +44 20 7991 5022

down further, the correlations within risk-on and risk-off asset groups have risen. Despite this, the overall level of cross asset correlations remains quite modest. This suggests that there is not a single dominant factor driving asset classes at the moment. The HSBC dendrogram analysis identifies three clear clusters of assets: (i) risk-on, (ii) USD-bloc, and (iii) rates. Interestingly, the rates cluster contains two sub-clusters with IG credit falling into one cluster and 10y sovereign bonds falling into the other.

Volatility Throughout 2019 we have seen an intense period of volatility compression across asset classes. This has been discussed at great length by many in the market. The usual focus of these discussions is a search for an answer to the question of why volatility is so low. We focus instead on the increase in gap risk since the end o f the financial crisis. This is something we discuss in detail in The Great Gapsby , 1 May 2019.

Disclosures & Disclaimer This report must be read with the disclosures and the analyst certifications in the Disclosure appendix, and with the Disclaimer, which forms part of it.

Issuer of report: HSBC report: HSBC Bank plc iew HSBC Global Research at:

https://www.research.hsbc.com

Trusted by over 1 million members

Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.

























Multi-Asset ● Quant 2 May 2019

Contents Taking a dip in the data

3

Goldilocks waning; correlations rising

3

Much ado about low vol

8

The Great Gapsby

9

Deep dive

11

Related reports

12

Asset returns

13

Monthly returns (sorted by performance)

14

Monthly returns (sorted by asset)

15

Annual returns (sorted by performance)

16

Annual returns (sorted by asset)

17

Risk indicators

18

Cross asset volatility indicator (CRAVIN)

19

HSBC risk monitor

20

Multi asset volatility indicator suite (MAVIS)

21

Asset class realised volatility histories

22

Asset class volatility monitors

26

Bond/equity portfolio volatility

40

Volatility DNA profile

41

Volatility waves

41

Correlations

42

Cross-asset correlations

43

Equity country correlations

46

Global equity sector correlations

49

US equity sector correlations

52

Europe equity sector correlations

55

Japan equity sector correlations

58

EM equity sector correlations

61

FX correlations

64

2-year rates correlations

67

10-year rates correlations

70

Commodities correlations

73


























Taking a dip in the data

Goldilocks waning; correlations rising Multi-asset performance during 2019 has been a pleasant change from a very challenging 2018. All major asset classes have performed well so far this year (Chart 1), and this has been accompanied by a widespread perception that we are now in a so-called Goldilocks period for financial markets.

1. Monthly asset-class total returns

Source: MSCI, Refinitiv Datastream, Bloomberg, HSBC Note: DM equity = MSCI World Index; EM equity = MSCI EM Index; DM rates = Global Treasury Index; DM IL = Global Aggregate Inflation-Linked Index; USD IG = US Investment Grade Corporate Credit Index; USD HY = US High Yield Corporate Index; EMD USD = EM USD debt; EM Local = EM local-currency debt; Comdty = Composite Commodity Index; REITs = Real Estate Investment Trusts Index. All returns are total returns in USD.


























such as Developed Market rates. Nevertheless, April has been a much more challenging environment than the beginning of the year. This risk-on move has been accompanied by a tick up in the HSBC RORO Indicator (Chart 2), which measures the strength of cross-asset correlations. This suggests to us that we might be seeing a waning of the Goldilocks story and are now simply living through a classic risk-on surge.

2. Strength of correlations across a wide range of asset classes

Source: Bloomberg, Refinitiv Datastream, HSBC

What does this mean for diversification?

When the RORO Indicator is at high levels it typically means that a single factor is dominating the returns of all asset classes. This, clearly, makes diversification challenging to achieve. However, whilst the indicator has recently ticked up, the overall level is still low – particularly relative to the levels seen since the financial crisis. In such a situation, interpreting what the recent increase in correlation strength means for diversification opportunities is a bit more subtle. We first look at what has happened to correlations for risk-on and risk-off asset groups separately. Since we are looking at a very recent increase in the RORO indicator, we look at short-term correlations (Chart 3). These correlations for risk-off assets appear to still be rising, whereas there are signs that the correlations amongst risk-on assets might have peaked. 3. Correlations across risk-on and risk-off asset cla sses 1.0 0.8 0.6 0.4 0.2 0.0 2001

2004

2 0 07

2 0 10

2013

2016

Risk-on correlation (rolling 13-week)

2 01 9

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 2 0 01

2 0 04

2 0 07

2010

2013

2 0 16

Risk-off correlation (rolling 13-week)

2019


























These embryonic signs of diversification amongst risk-on assets are bolstered by the evidence from the more traditional channel of multi -asset diversification – namely between risk-on and risk-off asset classes. The correlation between equity and bond returns in major d eveloped markets has been reliably negative since at least the start of the year (Charts 4 and 5) and has recently decreased further.

4. Bond-equity correlation (US)

Source: MSCI, Refinitiv Datastream, HSBC Note: Correlations calculated using weekly total returns over 6-month window

5. Bond-equity correlations (Germany)

Source: MSCI, Refinitiv Datastream, HSBC Note: Correlations calculated using weekly total returns over 6-month window

Drilling down into cross-asset correlations

With the RORO Indicator being at relatively modest levels it suggests that there is not just one dominant factor driving asset class returns at the moment. At times like this we find it useful to turn to our dendrogram analysis. These dendrograms are very information-rich charts which illustrate how different asset classes are moving together. They are produced by an




























Towards the right of the chart we can see (in red) a classic risk-on grouping of assets such as equities.



Towards the middle of the dendrogram (in blue) we can see what appears to be a USDdominated cluster of USD-FX rates and commodities. To us, it seems likely that the driving force behind this USD-dominated cluster is the dovish pivot from the Fed seen early this year.



Towards the left of the chart (in purple) we can see a DM rates cluster. There are two interesting points to note about this DM rates cluster, which we outline i n more detail below.

6. Multi-asset dendrogram


First, when we look within the cluster we can see that the three IG credit series have formed a cluster. This is unusual – we more typically see the USD IG corporate credit returns cluster most strongly with US 10y government bond returns, and similarly for the other currencies. This suggests that credit spreads have been much more closely correlated lately than usual. Indeed, Chart 7 shows that correlations among the three IG credit series have risen notably lately.

7. IG credit spreads much more closely correlated recently

0.8 0.6 0.4 0.2




























Second, whilst we would typically think of DM rates as being a classic risk-off asset class we can see that other classic risk-off assets such as the JPY and Precious Metals have been grouped into a rather strange-looking cluster with MXN and EM debt (local currency). This grouping is so bizarre that we suspect that it is an anomaly caused by a rapidly changing correlation structure. We calculate these correlations over a 6-month window: This window includes the sharp risk-off period towards the end of 2018, the classic Goldilocks moves of early 2019, and the most recent ‘classic’ risk-on move. It is likely that this 6 -month window is simply averaging over too many correlation regimes. Synchronised again – again – this this time on the downside

What the dendogram in Chart 6 is also showing is a tight cluster of 10 -year developed market (DM) sovereign bonds. This suggests a common, global driver behind the recent moves in longterm DM rates. From a fundamental point of view the most obvious driving force is the fairly synchronised slowing of global growth and inflation. Granted, some of the Q1 GDP prints were not as bad as commonly feared. But this was large ly growth ’ (30 Apr 19). This has due to one-offs and, as our economists put it, ‘The wrong type of growth’

been – and indeed is still being – reflected on the long long-end -end of sovereign bond yield curves almost across the board. As a result, correlations among 10-year sovereign bonds have risen strongly since the start of the year (Charts 8 and 9). We n ote Russia, Turkey and South Africa are simply outliers facing acute idiosyncratic challenges.

8. 10y rates correlations during 2018…

9. … and 2019 year -to-date -to-date

Source: HSBC, Bloomberg



























10. 2y rates correlations during 2018…

11. … and 2019 year -to-date -to-date



This highly correlated move YTD has not been confined to the 10-year segment though. Charts 10 and 11 show the same correlation matrices as above, but this time for 2 -year sovereign bond yields. Again, the important thing here is to compare the overall shading within the heat map, rather than the ordering of assets within it. This part of the yield curve reflects central bank expectations, rather than the outlook for (structural) growth and inflation. Despite the different natural drivers of 2y and 10y rates, we see a similar story. 2018 saw fairly weak correlations across the short-end of global sovereign bonds. During 2018 there were strong policy divergences between different countries. Some markets were pricing rate hikes, whereas others were still pricing cuts. This correlation profiles h as changed notably now. Chart 11 again exhibits much more red, reflecting the widespread and synchronised dovish dovish central bank pivot in the past four months. However, it is notable that the shift in correlations at the 2y point is not as stark as seen for the 10y point. This also underlines that the synchronised slowdown slowdown theme is well baked into longer dated rates, while the short-dated yields appear to be still linked with respective Central Bank’s rhetoric. There are still some rate hikes being priced into the front end of EUR and GBP curves while for the USD markets are categorically looking for cuts from here on. A subtle element of policy divergence, at least in the f orward rates markets, still remain.

Much ado about low vol


























12. The volatility wave has ebbed


Throughout 2019 we have seen an i ntense period of volatility compression. This can be clearly seen in our CRoss-Asset Volatility INdicator (CRAVIN, Chart 13). Market participants seem to demand precisely one explanation whenever we see periods of sustained low volatility. However, we rarely believe that low volatility environments can be explained by any single factor, believing instead that there are a range of possible explanations with varying degrees of plausibility (see our previous notes on cross-asset volatility: The Bonfire of the Volatilities, 25 Sep 2017 and Ideas for Jekyll or Hyde , 17 Jul 2014). We believe that the

two important factors behind the recent drop i n volatility are the dovish pivot of central banks and the consensus belief in a Goldilocks scenario.

13. CRAVIN on the back foot again 4.1 1.2 l e v e l r o t a c i

d n

0.6

0.0

95th percen percentile tile HSBC CrossCross-Asset Asset Volatility Indicator


























This gap risk is not just manifested in these spikes in implied volatility. It is sometimes even more pronounced for intraday price moves. In Charts 14-16 we compare the distribution of daily ranges for two time periods: (i) 2001-20082 and (ii) 2011-today3. Chart 14 shows the results for the S&P 500, Chart 15 for EUR-USD, and Chart 16 for 10Y Bund futures. 14. S&P 500 distribution of daily high-low

15. EUR-USD distribution of daily high-low

400

700 10

350 300 250

20 10

400

0

0

300

4.0% 4.8% 5.6% 6.4% 7.2% 8.0%

150

30

500

5

200

40

600

2%

3%

4%

200

100

100

50

0

0 0%

1%

2%

3%

4%

2001-2008

5%

6%

0.0% 0.0% 0.5% 0.5% 1.0% 1.0% 1.5% 1.5% 2.0% 2.0% 2.5% 2.5% 3.0% 3.0% 3.5% 3.5% 4.0% 4.0% 4.5% 4.5%

7%

2011-today

2001-2008


2011-today

Source: HSBC, Bloomberg;

15. 10Y Bund futures distribution of daily high -low 500

5

400 300

0 1.90

200

2.30

2.70

3.10

3.50

100 0 0.0

0.2

0.4

0.6

0.8

1.0

2001-2008

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

2012-today

Source: Bloomberg, HSBC

In all three cases there is more weight in the right tail of the distribution. This is particularly notable for the 10Y Bund future. It is interesting to see that the frequency of right-tailed moves are higher for Bunds than the S&P and EUR-USD. This could be partly due to the increased occurrences of politically driven safe haven flows into Bunds. Additionally, Bunds have gone through a sustained period of collateral squeeze during ECB QE. When the funding ma rket


























Deep dive Hopefully, like Goldilocks, you have found this first section to be “just right”. If, however, you feel that there is a gap in your knowledge which can only be filled by a multi-asset quant chart pack, then we have a wealth of information for you in the following pages. The deep dive begins with a look at returns from a variety of asset classes (page 13). 13). We follow this with an in-depth look at volatility (page 18) (page 18).. We analyse volatility not only within individual asset classes, but also from a multi-asset perspective with our HSBC CRAVIN (CRoss-Asset Volatility INdicator) indicator. Next we apply our powerful correlation analysis to a variety of

different data sets (page 42) (page 42).. First we investigate how different asset classes have moved together (page 43) (page 43),, then we focus on the correlation structure within single asset classes. Last, but not least, we track the correlation between sovereign bonds and equity markets (page 76). 76).




























Related reports

Previous Data Matters Data Matters: Diversification… apart from when you need it , 17 January 2019 Data Matters: RORO Resurgence , 17 October 2018 Data Matters: EM-DM Vol Divergence , 8 August 2018 Data Matters: EM underperforms, US dominates , 23 May 2018 Data Matters: Riddles, mysteries and enigmas , 12 April 2018

Other periodicals HSBC Multi-Asset Heat Map (interactive webpage) HSBC Surprise Indices , 1 May 2019 HSBC FX Positioning Indicators , 29 April 2019


























Asset returns

Section contents Monthly returns (by performance)

p. 14

Monthly returns (by asset)

p. 15

Annual returns (by performance)

p. 16

Annual returns (by asset)

p. 17

























Monthly returns (sorted by performance) 1. Monthly asset-class total returns

Source: MSCI, Refinitiv Datastream, Datastream, Bloomberg, HSBC

M

Note: DM equity = MSCI World Index; EM equity = MSCI EM Index; DM rates = Global Treasury Index; DM IL = Global Aggregate Inflation-Linked Index; USD IG = US Investment Grade Corporate Credit Index; USD HY = US High Yield Corporate Index; EMD USD = EM USD debt; EM Local = EM local-currency debt; Comdty = Composite Commodity Index; REITs = Real Estate Investment Trusts Index. All returns are total returns in USD.

u lt iA s s 2 e M t a

●

y

Click for:

14

Contents | Asset returns | Risk indicators | Correlations

Q 9

1

0

2 t

n

a

u



















Monthly returns (sorted by asset) 2. Monthly asset-class total returns

Source: MSCI, Refinitiv Datastream, Bloomberg, HSBC

M

Note: DM equity = MSCI World Index; EM equity = MSCI EM Index; DM rates = Global Treasury Index; DM IL = Global Aggregate Inflation- Linked Index; USD IG = US Investment Grade Corporate Credit Index; USD HY = US High Yield Corporate Inde x; EMD USD = EM USD debt; EM Local = EM local-currency deb t; Comdty = Composite Commodity Index; REITs = Real Estate Investment Trusts Inde x. All returns are total retu rns in USD.


●

y

Click for:


Q 2 0 1 9

15

t

n

a

u



















Annual returns (sorted by performance) 3. Annual asset-class total returns over the past 25 years


M

Note: DM equity = MSCI World Index; EM equity = MSCI EM Index; DM rates = Global Treasury Index; DM IL = Global Aggregate Inflation-Linked Index; USD IG = US Investment Grade Corporate Credit Index; USD HY = US High Yield Corporate Index; EMD USD = EM USD debt; EM Local = EM local-currency debt; Comdty = Composite Commodity Index; REITs REITs = Real Estate Investment Trusts Index. All returns are total returns in USD.


●

y

Click for:

16


Q 9

1

0

2 t

n

a

u





















Annual returns (sorted by asset) 4. Annual asset-class total returns over the past 25 years


Note: DM equity = MSCI World Index; EM equity = MSCI EM Index; DM rates = Global Treasury Index; DM IL = Global Aggregate Inflation-Linked Inflation-Linked Index; USD IG = US Investment Grade Corporate Credit Index; USD HY = US High Yield Corporate Index; EMD USD = EM USD d ebt; EM Local = EM local-currency debt; Comdty = Composite Commodity Index; REITs = Real Estate Investment Trusts Index. All returns are total returns in USD.

u

M


























Risk indicators

Section contents Cross asset volatility indicator (CRAVIN)

p. 19

Risk monitor

p. 20

Multi asset volatility indicator suite

p. 21

Asset class realised volatility histories

p. 22


























Cross asset volatility indicator (CRAVIN) 5. Composite index of 3-month realised volatility across a sset classes



























HSBC risk monitor

6. Cross-asset risk premia


























Multi asset volatility indicator suite (MAVIS) 7. Annualised realised volatility ranges across ass et classes




























Asset class realised volatility histories DM equities 8. Developed market equity total return index 3-month realised volatility


























Inflation-linked bonds 11. Inflation-linked bond total return index 3-month realised volatility


























EM USD debt 14. Emerging market USD debt total return index 3-month realised volatility


























REITS 17. REITs total return index 3 -month realised volatility


























Asset class volatility monitors Developed market country equity volatility 19. Developed market country equity realised volatility ranges

























































































































































































































































































































































































































































































































































































































































































Emerging market equity sector correlation indicator i ndicator 63. Indicator measuring the strength of EM equity sector correlations

Source: MSCI, Refinitiv Datastream, HSBC

The correlation indicator measures the strength of the correlations between the assets listed in the table on the previous page. An increase in the indicator implies rising correlations, with a value of one corresponding to perfect correlations between all assets. See page 78 for more details.

Key emerging market equity sector correlations 64. Dendrogram showing the key EM equity sector correlations over the past 6 months

Source: MSCI, Bloomberg, Refinitiv Datastream, HSBC


























Major changes in EM equity sector correlations 65. EM equity sectors where the correlations have changed significantly over over the past 12 months






























FX correlations 66. USD-based FX correlations over the past 6 months

Source: Bloomberg, HSBC Note: The assets are ordered such that those with similar correlations to other assets are next to each other.

The heat map shows the correlations between different assets over the p ast six months. 

Dark red regions indicate strong positive correlations;



Dark blue regions indicate strong negative correlations;



Yellow/green indicates weak correlation/uncorrelated assets.

Table 10. USD-based exchange rates included in the heat map


























FX correlation indicator 67. Indicator measuring the strength of correlations between USD-based exchange rates



Key FX correlations 68. Dendrogram showing the key USD-based FX correlations over the past 6 months



























Major changes in FX correlations 69. USD-based FX rates for which the correlations have changed significantly over the past 12 months
































2-year rates correlations 70. 2-year interest rate swap correlations over the past 6 months


The heat map shows the correlations between different assets over the p ast six months. 



























2-year rates correlation indicator 71. Indicator measuring the strength of correlations between 2-year interest rate swaps



Key 2-year rates correlations 72. Dendrogram showing the key 2- year swap rate correlations over the past 6 months


























Major changes in 2-year 2- year rates correlations 73. 2-year swap rate pairs for which correlations have changed significantly over over the past 12 months






























10-year rates correlations 74. 10-year interest rate swap correlations over the past 6 months


The heat map shows the correlations between different assets over the past six months. 



























10-year rates correlation indicator 75. Indicator measuring the strength of correlations between 10-year interest rate swaps



Key 10-year rates correlations 76. Dendrogram showing the key 10- year swap rate correlations over the past 6 months




























Major changes in 10-year rates correlations 77. 10-year swap rate pairs for which correlations have changed significantly over over the past 12 months




























Commodities correlations 78. Commodities correlations over the past 6 months


























Commodities correlation indicator 79. Indicator measuring the strength of correlations between commodities



Key commodities correlations 80. Dendrogram showing the key commodities correlations over the past 6 months


























Major changes in commodities correlations 81. Commodity pairs for which correlations have changed changed significantly over the past 12 months




























Bond/equity correlations US 82. US 10Y government bonds and equities correlation


Note: Correlations calculated calculated using weekly total returns for bonds and price returns for equities over 6m windows.

Japan 83. Japan 10Y government bonds and equities correlation




























Germany 85. Germany 10Y government bonds and equities correlation


Note: Correlations calculated calculated using weekly total returns for bonds and price returns for equities over 6m windows.


























Methodology

Correlation indicators The various correlation indicators in this report provide aggregate measures of the level of correlations between broad groups of assets. In each case, the starting point is a set of total returns4 (or yield differences in the case of rates). We then use principal component analysis (PCA) to decompose these returns into a series of principal components (PCs), which are a new set of series that represent the common movements movements across the original time series. The first principal component can essentially be thought of as the most important factor driving


























Alternatively, you could specify specify how dissimilar dissimilar objects within a cluster can be and and then choose the hierarchical clustering which results at this distance level. Another option would be to attempt to split the hierarchy according to some systematic rule based on the distances at which different clustering arrangements occur. Defining similarity

In order to decide which two objects should be joined together as the first non-unitary cluster, first you need to define a distance between each pair of objects. There are various ways of doing this; we use the following, somewhat strange-looking definition:

 = √ 2(1− 2 (1− ) , where  is the distance between objects  and  and  is the correlation between objects  and  This definition is not as bizarre as it might initially appear and is strongly linked to the Euclidean distance between the two time-series5. Making a new cluster

To create the first cluster, the smallest value of d ij is found; objects i and and j are are then joined into a ij is


























the distance between cluster (A,B) and cluster C is 0.75, which can be seen in Table 15. The distances between (A,B) and the remaining clusters are calculated similarly. In Table 15 we have highlighted the smallest distance: 0.2 between clusters C and D. We thus join these two into a new cluster cluster (C,D) and calculate the distances between this new cluster and all others. This gives the new distance matrix shown in Table 16.

Table 16. After combining C and D into (C,D) (A,B) (C,D) E

(A,B) 0 0.85 0.325

(C,D)

E

0 0.45

0

Source: HSBC

In Table 16 we can see that the d istance between cluster (A,B) and cluster (C,D) is 0.85. Where did this number come from? It is the average of the distances between: 

A and C (0.8)


























88. Average linkage clustering dendrogram




























change as new data become available and the means and standard deviations are updated. It’s

easy to show though that provided the initial window is reasonably long, the choice of this window doesn’t have much of an effect on the final indicator.

Table 18. Assets included in the cross asset volatility indicator Equities North America

Credit USD Investment Grade

Rates US 10Y

FX EUR-USD

Commodities Industrial Metals


























Disclosure appendix Analyst Certification

The following analyst(s), economist(s), or strategist(s) who is(are) primarily responsible for this report, including any analyst(s) whose name(s) appear(s) as author of an individual section or sections of the report and any analyst(s) named as the covering


























Hold; when it is between 5% below and 5% above the current share price, the stock will be classified as a Hold; when it is between 5% and 20% below the current share price, the stock may be classified as a Hold or a Reduce; and when it is more than 20% below the current share price, the stock will be classified as a Reduce. Our ratings are re-calibrated against these bands at the time of any 'material change' (initiation or resumption of coverage, change in target price or estimates). Upside/Downside is the percentage difference between the target price and t he share price.


























Distribution of fundamental credit and covered bond recommendations

As of 01 May 2019, the distribution distribution of all independent fundamental credit credit recommendations published published by HSBC is as follows: All Covered issuers

Overweight Neutral Underweight

Issuers to whom HSBC has provided Investment Banking in the past 12 months

Count

Percentage

Count

Percentage

93 227 88

23 56 21

48 104 15

52 46 17


























MSCI Disclaimer

The MSCI information included in this report is for your internal use only, may not be reproduced or redisseminated in any form and may not be used as a basis for or a component of any financial instruments or products or indices. None of the MSCI information is intended to constitute investment advice or a recommendation to make (or refrain from making) any kind of investment decision and may not be relied on as such. Historical data and analysis should not be taken as an indication or guarantee of any future performance analysis, forecast or prediction. The MSCI information is provided on an "as is" basis and the user of this information assumes the entire risk of any use made of this information. MSCI, each of its affiliates and each other person involved in or related to compiling, computing or creating a ny MSCI information (collectively, the "MSCI Parties") expressly




























Data Matters - HSBC.PDF

Recommend Documents