1 / 23

September 2019

September 2019. How strategies need to adapt over time Active selection and deployment. Dr Jens Kroeske, PRM FRM Head of Macro Systematic Strategies Research Investment Innovation. Active Systematic Investing.

lok
Download Presentation

September 2019

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. September 2019 • How strategies need to adapt over time • Active selection and deployment • Dr Jens Kroeske, PRM FRM • Head of Macro Systematic Strategies Research • Investment Innovation • Active Systematic Investing • For limited professional use only – Not for general distribution. The following presentation outlines a fund concept presently under internal consideration with the intention of obtaining preliminary feedback on such a product. The fund concept is not available for sale and any product details discussed are reflective of current market conditions and are subject to change • Aberdeen Standard Investments is a brand of the investment businesses of Aberdeen Asset Management and Standard Life Investments

  2. Active Systematic Investing Active systematic – an oxymoron? • All quantitative modelling involves choices • Macro systematic – extension to other asset classes • Importance of ‘owning the model’ Should systematic strategies adapt over time? • Quality control dictates supervision of models • The fear of auto-pilot strategies • Bayesian approach: have a prior, update your believes 3. How to control the outcome • Introducing dimensionality • Portfolio construction, make it outcome oriented • Scenario analysis Markets are adaptive systems that require both experience and qualitative judgement to successfully navigate

  3. Active systematic – an oxymoron?

  4. Active Systematic Investing - All quantitative modelling involves choices • The fable of factor investing • Quants tend to ‘hide behind their models’ • Source: Bloomberg, Aberdeen Standard Investments. For illustrative purposes only. No assumptions regarding future performance should be made • Systematic investing is mostly taken to be an extension of factor investing in long only equity factors (value, momentum, etc.) But even traditional equity factor strategies involve choices: • Definition of the factors - parameters • Long/short stocks vs Long stocks/short benchmark • Beta adjustment • Orthogonalisation Referring to risk (percentage of variance explained) does not define factors • Need a view on expected return • Rotational indeterminacy • This results in significantly differing outcomes for investors

  5. Macro Systematic Investing • The fallacy of consistency • Extension to other asset classes is non-trivial • Consistency of definition or different market structure? • Is the asset grid (asset class and style) really the benchmark? • How does it relate to outcomes? • In each case it is necessary to take a view on a model • All systematic investing has an active component • Investment outcomes depend heavily on these active choices • Outcome oriented systematic investing • View systematic strategies as tools in a toolbox, not investment outcomes in themselves • Systematic investing is mostly about a systematic quantitative investment process and applying decision making rules consistently.

  6. Importance of ‘owning the model’: Implementation matters • Equity volatility risk premia • Do clients care about path dependency or tail risk? • Source: Aberdeen Standard Investments, Bloomberg. For illustrative purposes only. No assumptions regarding future performance should be made One of the best known alternative risk premia Strategy aims to benefit from the difference between implied and realised volatility in equity markets Vanilla example rolling straddles monthly The choice of when to roll the strategy can have a material impact on outcomes (Market structure and congestion) Takeaway points Trading experience of this market Careful strategy design Strong emphasize on robustness Focus on the outcome we are trying to achieve

  7. Should systematic strategies adapt over time? “When the facts change, I change my mind. What do you do, sir?” John Maynard Keynes

  8. What is more expensive? • Just because valuation models on volatility selling are less established, it does not mean that they do not matter • Source: Thomson Reuters DataStream, Bloomberg. Aberdeen Standard Investments. For illustrative purposes only. No assumptions regarding future performance should be made Equity valuations • PE (trailing/forward, CAPE) • PB • Relative to own history, relative to peers Volatility selling valuations • Implied volatility • VRP = implied – realised volatility • Relative to own history, relative to peers

  9. The fear of auto-pilot strategies • Source: Barclays Live, Aberdeen Standard Investments • Credit carry without carry • Average carry above costs on average: 10bp • Back-test looks attractive • Carry above costs today: 2bp (end of July) • Single quantitative investment strategy providers have no incentive to recommend exiting a strategy • Can lead to unnecessary trading costs with little return potential • View systematic investment strategies as tools,not outcomes in themselves

  10. Quality control dictates supervision of models • 73% of investors believe alternative risk premia strategies cannot be timed, but the percentage is falling (*) • Source: (*) MJ Hudson Allenbridge Alternative Risk Premia Fund Review 2019 • Systematic investing is based on the premise that excess returns can be earned because of: • Risk premia – rational compensation for taking risk in ‘bad times’ • Behavioural effects – momentum, over-reaction • Structural effects – leverage constraints, supply and demand imbalances, benchmark investing • Commonly accepted that risk premia, behavioural and structural effects are time varying • Most traditional risk assets have valuation models • Urgent need for valuation models for all risk premia models (carry, volatility selling, etc.) • Structural and behavioural effects similarly need screening: • Is the carry still there/sufficient? • Are demand/supply effects (structured product issuance) still dominating? • How do market trends (move to passive) affect behavioural strategies? – can you really wait 20y to be statistically sure?

  11. Markets evolve and so should we • Be Bayesian, not frequentist • Take a view on the model, not the market – there is no such thing as statistical certainty • Most people do not believe systematic strategies can be timed • Markets are too efficient • Sparse data availability can lead to ‘overfitting to the recent past’ • Quantitative investors are often reluctant to ‘take a view’ on the markets. But they are paid to ‘take a view on the model’ • But frequentist approach is inappropriate (p-hacking) • Take a Bayesian approach: have a prior and update believes when new data comes in • Market structures do evolve (e.g. Bretton Woods, inflation targeting, quantitative easing, japanification) • Definitions of valuations change (P/B, etc) • Investor segmentation shifts (DB to DC) • Regulations change (Dodd-Frank, Libor discontinuation)

  12. How to control the outcome?

  13. What to take from a back-test? • Again, be Bayesian, not frequentist • Back-testing is ubiquitous in systematic investing • What is a realistic Sharpe ratio? • In the absence of much relevant data, revert to prior • Update your believes as more data becomes available • All performance is sensitive to parameter choices, but also.. • All performance is conditional on the market environment (QE, rise of globalisation). • Always ask the questions: • Is the strategy really unaffected by macro forces (growth, inflation, monetary policy)? • How to react if the sensitivity changes over time? • Adapt the strategy • Lower expected diversification potential

  14. Portfolio construction • Portfolio construction should be done with an outcome in mind • Methods like risk parity, have no direct link to outcomes on the portfolio level beyond their intuitive appeal and (perhaps lucky) historical success • ‘Diversify across everything’ can easily mislead investors into believing that there are a great number of independent return drivers, but that is unrealistic • Be wary of off-setting risks reducing return potential • Negatively correlated strategies can increase leverage for volatility targeting funds • Relying on negative correlations in crisis is challenging unless it is an explicit hedge • Managers are paid to take exposures, not offset them elsewhere in the portfolio • Always be clear to clients what can realistically be achieved in terms of diversification • Investors are disappointed to see strategies suddenly becoming correlated • What drawdown risk is realistic, in particular with high Sharpe ratio, negatively skewed payoffs?

  15. Introducing Dimensionality • Source: Thomson Reuters Datastream, Bloomberg, Aberdeen Standard Investments • How many independent drivers are there really? – Call them dimensions • Traditional asset classes have limited number of drivers • How much more do systematic strategies offer? • Traditional asset classes • Traditional asset classes only contain two main dimensions: Equity-like and duration-like exposures • Portfolio construction can improve dimensionality, but to go beyond 2-3 dimensions, a wider investment universe is needed Dimensionality US institutional median portfolio 1.4 1.2

  16. Dimensionality approach – Built on years of academic research • In collaboration with University of Edinburgh • Full-time academic resource integrated into Macro Systematic Investing team

  17. Our solution – ‘Dimensions’ approach • Academically validated approach to superior portfolio construction • In 2013 our industry-renowned team of risk professionals launched a project to investigate a new approach to portfolio construction based on finding truly independent return streams within a portfolio that we now call Dimensions. • The greater the number of Dimensions you can create in you portfolio the better. However, in practice we found FOUR were the largest number one could reliably create. • Definition • “A dimension is a group (or cluster) of strategies that behave independently from the rest of the portfolio”

  18. The Power of 4 Dimensions • Doubling risk-adjusted return, quartering the tail risk • Source: Aberdeen Standard Investments. Example uses uncorrelated assets each with a Sharpe ratio of 0.5 and high fat tail risk Higher Sharpe ratios 0.5 0.5 Create 4 uncorrelated groups of strategies each with a Sharpe ratio of 0.5 Lower kurtosis ‘fat tails’ 0.5 0.5 Portfolios that reach a dimensionality level of four… …deliver a better investment outcome… …more consistently

  19. All models are wrong, but some are useful • All quantitative investing should be evidence based and model driven • But combining individual systematic strategies into a portfolio is a hard problem • Return drivers can change over time and overlap in unexpected ways • Short volatility, short JPY, short duration • Long carry, long mean reversion • Most investors still want experienced portfolio managers to oversee quantitative portfolios • This is mostly the case for risk avoidance that can/should be discretionary • Not much data for 1 in 200y events • Scenario analysis is key

  20. Forward looking risk analysis: Proprietary scenario engine • Better to be approximately right than exactly wrong Scenario analysis is used to explore tail risks • Standard risk models are focused on the middle of the distribution • Scenarios should focus on potential portfolio weaknesses, not spurious accuracy • Penalize or exclude strategies with relatively high perceived future downside risk • We look at different types of scenarios: • Historical scenarios • Large scenario library • Creation of bespoke scenarios • Forward-looking scenarios • Internally modelled • Extreme, but economically plausible • Guided by qualitative expert judgement • Non-traditional stress scenarios • Derivative positions • Focus on nonlinear and convex return profiles Historical Forward-looking Representative portfolio Source: Aberdeen Standard Investments, 31 July 2018 This is intended for illustrative purposes only.

  21. Conclusion • Aberdeen Standard Investments is a brand of the investment businesses of Aberdeen Asset Management and Standard Life Investments.

  22. Conclusions • There is a place for ‘active’ in systematic • All quantitative modelling involves active choices • Take a Bayesian approach to investing, not a frequentist one • Have a prior, update your believes • Focus on outcomes – systematic strategies are tools • Active risk avoidance is needed when models fail

  23. GR-DIS-001 311013 For professional clients only – Not for public distribution • Past performance is not a guide to future results. The value of investments, and the income from them, can go down as well as up and clients may get back less than the amount invested. • Aberdeen Standard Investments is a brand of the investment businesses of Aberdeen Asset Management and Standard Life Investments • The views expressed in this presentation should not be construed as advice or an investment recommendation on how to construct a portfolio or whether to buy, retain or sell a particular investment. The information contained in the presentation is for exclusive use by professional customers/eligible counterparties (ECPs) and not the general public. The information is being given only to those persons who have received this document directly from Aberdeen Asset Managers Limited or Standard Life Investments Limited (together “Aberdeen Standard Investments”) and must not be acted or relied upon by persons receiving a copy of this document other than directly from Aberdeen Standard Investments. No part of this document may be copied or duplicated in any form or by any means or redistributed without the written consent of Aberdeen Standard Investments. • The information contained herein including any expressions of opinion or forecast have been obtained from or is based upon sources believed by us to be reliable but is not guaranteed as to the accuracy or completeness. • Any data contained herein which is attributed to a third party ("Third Party Data") is the property of (a) third party supplier(s) (the “Owner”) and is licensed for use by Standard Life Aberdeen*. Third Party Data may not be copied or distributed. Third Party Data is provided “as is” and is not warranted to be accurate, complete or timely. To the extent permitted by applicable law, none of the Owner, Standard Life Aberdeen* or any other third party (including any third party involved in providing and/or compiling Third Party Data) shall have any liability for Third Party Data or for any use made of Third Party Data. Neither the Owner nor any other third party sponsors, endorses or promotes the fund or product to which Third Party Data relates. • Germany: Aberdeen Asset Managers Limited, registered in Scotland (SC108419) at 10 Queen’s Terrace, Aberdeen, AB10 1XL. Standard Life Investments Limited. Registered in Scotland (SC123321) at 1 George Street, Edinburgh EH2 2LL.  Both companies are authorised and regulated in the UK by the Financial Conduct Authority.  • * Standard Life Aberdeen means the relevant member of Standard Life Aberdeen group, being Standard Life Aberdeen plc together with its subsidiaries, subsidiary undertakings and associated companies (whether direct or indirect) from time to time. • Aberdeen Asset Managers Limited is registered in Scotland (SC108419) at 10 Queen’s Terrace, Aberdeen, AB10 1XL, Standard Life Investments Limited is registered in Scotland (SC123321) at 1 George Street, Edinburgh EH2 2LL, and both companies are authorised and regulated in the UK by the Financial Conduct Authority. • GB-040919-98303-1

More Related