Nomura Holdings, Inc. · TSE:8604
Nomura — BNP Paribas Asset Management — Nomura Global Quantitative Equity Conference 2014
18 slides from Nomura's conference presentation, May 8, 2014
Slide 1: Inter-temporal risk parity
FOR PROFESSIONAL INVESTORS Inter-temporal risk parity Nomura Global Quantitative Equity Conference 8th May 2014
Slide 2: Inter-temporal risk parity strategy
2 Inter-temporal risk parity strategy ● What is inter-temporal risk parity? – Systematic strategy rebalancing between a risky asset and cash – Weight of risky asset is chosen so that ex-ante risk is kept constant 𝜅 𝜅 𝑟𝐼𝑅𝑃 𝑡 = 𝑟𝑡 + 𝑟𝑐 (1 − ) 𝜎𝑡 𝜎𝑡 𝑟𝐼𝑅𝑃𝑡 performance of the…
Slide 3: Inter temporal risk parity strategy in practice
3 Inter temporal risk parity strategy in practice 25% I-GARCH Target (k) 20% Volatility 15% 10% Increase in volatility 5% 0% 80% One day lag for implementation Exposure of risky asset Exposure of cash 70% 60% Exposure 50% Decrease the exposure 40% 30% 20% 31-Dec-13 11-Feb-14 25-Feb-14 08-Apr-14 22-A
Slide 4: If returns of risky assets had Gaussian distributions
4 If returns of risky assets had Gaussian distributions Inter-temporal Gaussian distributed returns Buy and Hold Risk Parity Average annualized excess return 7.5% 7.7% Average annualized volatility 18.8% 19.3% Sharpe ratio 0.40 0.40 Maximum drawdown (MDD) -37.6% -38.5% Ratio MDD / volatility -2.0 -2
Slide 5: Inter-temporal risk parity applied to equities
5 Inter-temporal risk parity applied to equities ● Evidence that managing equities at constant risk adds value: – Hocquard, Ng and Papageorgiou (2013) – Cooper (2010) – Kirby and Ostdiek (2012) Higher Sharpe ratio and smaller drawdowns with constant volatility portfolio.. – Ilmanen & Kizer (2012) –
Slide 6: Understanding inter-temporal risk parity strategies
6 Understanding inter-temporal risk parity strategies Monte Carlo simulations with scenarios generated from parametric models 1 𝑟𝑡 = 𝜇 + 𝜎𝑡 𝑧 ● Apply different stochastic models [1] for risky asset returns – Keep risk premium m constant over time 2 𝜎𝑡 2 = 𝜔 + 𝛼 𝑟𝑡 − 𝜇 2 + 𝛽𝜎𝑡−1 2 ●…
Slide 7: Volatility clustering explains better risk-adjusted performances
7 Volatility clustering explains better risk-adjusted performances Using standard GARCH model Buy and Inter- GARCH with a = 9% and = 90% temporal Hold ● Generate volatility clustering while keeping risk Risk Parity premium is constant: Average annualized excess return 7.5% 9.1% – Higher Sharpe rat
Slide 8: Fat tails, leverage effect and skew
8 Fat tails, leverage effect and skew Inter-temporal GARCH with t-Student noise Buy and Hold Risk Parity ● Fat tails (GARCH with t-student noise) Average annualized excess return 7.4% 10.3% – Increase the probability of extremes events Average annualized volatility 17.8% 18.8% Improvement of the Sh
Slide 9: Impact of rebalancing frequency
9 Impact of rebalancing frequency 0.10 Improvement in Sharpe ratio 0.08 0.06 0.04 0.02 0.00 Daily Monthly Weekly Buy & Hold Semiannual Source: R Perchet, R Leote de Carvalho, T Heckel and P Moulin, “Inter-temporal risk parity: A constant volatility framework for equities and other asset classes” (20
Slide 10: Forecasting volatility (S&P500)
10 Forecasting volatility (S&P500) 12% GARCH* 12% NA-GARCH* 11% 11% ● Different GARCH models considered 10% 10% – Target 10% volatility ex-ante 9% 9% ● GARCH which includes 8% 8% 7% 7% – Volatility clustering Jan-91 Jan-94 Jan-97 Jan-00 Jan-03 Jan-09 Jan-12 Jan-91 Jan-94 Jan-97 Jan-03 Jan-06 Jan-09
Slide 11: Larger clustering effect in riskier asset classes
Larger clustering effect in riskier asset classes ● Estimation of clustering effect and fat tail US MSCI S&P US high US 10Y events for main asset classes Russell 1000 Emerging GSCI yield Invest. Grade Gov. Markets Commo. Bonds Bonds – Large a for Equities, in particular bonds Emerging, and for US hi
Slide 12: Returns with Inter-temporal risk parity strategy
12.0% -12.0% 0.0% 4.0% 8.0% -8.0% -4.0% 0.0% 2.0% 4.0% -4.0% -2.0% Jan-93 Jan-93 Jan-95 Jan-95 Jan-97 Jan-97 Jan-99 Jan-99 Jan-01 Jan-01 Russell 1000 (B&H) Jan-03 Jan-03 Jan-05 Jan-05 US 10Y Gov. Bonds (B&H) Jan-07 Jan-07 Jan-09 Jan-09 Jan-11 Jan-11 Jan-13 Jan-13 12.0% -12.0% 0.0% 4.0% 8.0% -8.0% -4
Slide 13: Historical simulations for different asset classes
Historical simulations for different asset classes US high US Invest. US 10Y ● Inter-temporal risk parity strategy Russell MSCI EM S&P GSCI yield Grade Gov. 1000 Markets Commo applied to equity indices and other asset Bonds bonds Bonds classes Buy and hold strategy Average annual. – Higher Sharpe ra
Slide 14: Factor investing
14 Factor investing ● Factor investing has been gaining attention since Fama & French (1992,1993) – Value and Size premiums in equity markets ● Carhart (1997) extended Fama and French model – Momentum premium was added ● Qian, Sorensen and Hua (2009) found value premium in other asset classes – Gove
Slide 15: Value and Momentum premiums
15 Value and Momentum premiums Inter-temporal risk parity strategy applied to Value and Momentum factors: ● Equities: daily data from Ken French’s web-site: – Value premium: HML (High-Minus-Low factor) – Momentum premium: Mom (Momentum) ● Sovereign Government bonds based on 10 countries*: – Value pr
Slide 16: Improvement of information ratios in factor investing
16 Improvement of information ratios in factor investing Momentum Value Momentum Momentum Value Fixed Applying inter-temporal risk parity to Equity Value Equity Foreign Exchange Foreign Exchange Fixed Income Income factor investing also bring benefits Buy and hold strategy ● Improvement of informati
Slide 17: Conclusions
17 Conclusions ● No Gaussian behavior or returns explains why constant volatility strategy add value ● Investors should think in terms of risk budget allocation rather than fixed weights ● Improvement of Sharpe ratio and information ratio explained by volatility clustering – Volatility is not consta
Slide 18: Disclaimer
18 Disclaimer This material is issued and has been prepared by BNP Paribas Asset Management S.A.S. (“BNPP AM”)*, a member of BNP Paribas Investment Partners (BNPP IP)** . This material is produced for information purposes only and does not constitute: 1. an offer to buy nor a solicitation to sell, n
BNP Paribas Asset Management — Nomura Global Quantitative Equity Conference 2014 on 6ix