20160614(화) - Japanese government bond yields fall to fresh low

http://www.ft.com/intl/cms/s/0/ac73458c-31d1-11e6-bda0-04585c31b153.html#axzz4BYiI2djN

“We think additional easing is only a matter of time as long as the BoJ adheres to its 2 per cent inflation target,” said Naohiko Baba at Goldman Sachs in Tokyo.

A range of tactical problems make it hard to act this week, however, even if the US Federal Reserve holds rates a day earlier. A dovish message from the Fed could push the yen up even further.

The BoJ’s biggest tactical obstacle is the Brexit referendum. If it eases this week and Britain then votes to leave the EU the yen might rise anyway. That risks fuelling a public perception that BoJ policy is ineffective.

Easing now also risks the appearance of political interference ahead of upper house elections on July 10, while the BoJ will not publish new economic forecasts — accounting for prime minister Shinzo Abe’s decision to postpone a scheduled rise in consumption tax — until its next meeting in July.

All of those factors make July a more likely date for action than June, although BoJ governor Haruhiko Kuroda likes to surprise.

Quant Hedge Fund Balance Computer and Brain Power

http://www.institutionalinvestor.com/article/3558232/asset-management-hedge-funds-and-alternatives/quant-hedge-funds-balance-computer-and-brain-power.html?ArticleId=3558232#/.V04BgWaYtRM.facebook


MAN-AHL

The firm trade four main strategies
1. classical long term trend following
2. proprieatory trend following
3. multistrategy
4. long only

- Man has been developing machine learning algorithms for half a decade.

"But the traders and scientists must know how the algorithms work and how avoid overfitting data to a trading hypothesis.

"You have to understand both the problem and the data, and fit them together carefully"
"You always run the risk of overfitting"


Aspect
- Aspect's trading program is actually seven in one, operating at a range of frequencies by capturing trends of between two weeks and six months.
- Aspect occupies a middle ground, blending human intuition and rigorous research to create trading programs that aren't just autonomous algo.

"Our research approach has always been hypothesis-driven"


WINTON
- Open a San Francisco data science center to tap Silicon Valley's talent
- See a future in building proprietary data sets.
- Winton ventures, a new venture capital unit that is on the hunt for data-driven start ups
- "We are interested in companies where really understanding what can be inferred from the data - drawing valid conclusion from the data - is essential to the success of the business

Directional or Cross-sectional - a matter of choice

https://www.man.com/2/unconventionalviews-2016-sargaison
2016.01.12


Quantitative investment
1. CTA
- much more directional approach to trading

2. Traditional quant equity business
- aim to be market neutral
- to neutralise exposure to the broad equity market by trading only cross-sectional variations in returns
- 'relative value'

과거를 살펴보면 평균적으로 비슷한 퍼포먼스. 하지만 둘의 상관관계는 낮음. (different driving force)
Both have been successful in different conditions over a long period.

어떤 포트폴리오를 구성하고 있는 자산들 간의 상관관계가 fully uncorrelate 상태가 아니라면 자산의 return 에 영향을 미치는 the largest common factor 가 있다고 생각 할 수 있다..

cross sectional 접근과 time-series 접근이 여기서 나뉘는데..
cross sectional 접근은 바로 이 largest common factor 를 제거 하는 것.
그럼 수익률이 소스는?
그 외 다른 smaller factor. 예를들면 Value.

최근에 하버드 연기금을 운용하는 구루인 Stephen Blyth가 말하길,
he felt current markets, with continuing high valuations, were becoming "frothy" and he further stated that they had now
"renewed focus on identifying... managers with demonstrable investment expertise on both the long and short sides of the market."

===>  CTA 가 하는일.

CTA는 지난 18년동안 cross-sectional 전략들과 비슷한 수익률을 가지고 있지만 두달 연속 10%이상 S&P 가 빠졌던 bear 장에서의 수익률은 cross-sectional 보다 훨씬 좋다. (annualized 5.1%)

In the context of a truly diversified portfolio, we believe neutralising all underlying market exposures is seldom the optimal approach, even during periods of extreme stress and acute risk aversions.

IS THERE ANY ALPHA LEFT IN FX?

https://www.man.com/2/is-there-any-alpha-left-in-FX
2015.01.30

In the coming few years, we believe there will be two sources of alpha in FX
1. while big trend have been elusive recently, they are likely to return as economies  start diverging again.
==> This process is already started as the US economy is outperforming

2. There are a number of alpha sources independent of big trends, such as macroeconomic fundamentals an non-trend statistical indicators.

-----------------------------------
EXCHANGE RATE TRENDS WILL RETURN AN ECONOMIC FUNDAMENTALS DIVERGE

1. Foreign exchange rates reflect the relative economic fundamentals of a pair of countries.

2. When two countries'(or regions) fundamentals diverge significantly over an extended period, the resulting adjustment in their exchange rate often leads to a multi-year trend.

3. Historically, most currency managers based their trading on momentum and carry strategies.

뒤에 붙어있는 References 들에 볼게 좀 있는듯..

꿀

http://www.thebondbeat.com/bondbeat/wp-content/uploads/

Research Paper 목록(작성중)

<Deutsche Bank>
1.QCD Model: DB Quant Handbook, 22 July 2010
2. Signal Processing : Quant Tactical Asset allocation, 19 Sep 2011
3. Signal Processing: New Insights in Country Rotation, 9 Feb 2012
4. Signal Processing: The Rise of the Machines, 5 June 2012
5. Signal Processing: Style Rotation, 7 Sep 2010
6. Signal Processing: Pairs trading with a fundamental flavor, 20 MAr 2012
7. Signal Processing: Portfolios Under Construction: Correlation & Consequences, 24 Jan, 2012
8. Signal Processing: Macro Uncertainty, Investor Sentiment, and Asset Returns, 30 May 2013

<MS>
1. US Quant Research: Agreeing to Disagree : Measurement Error in Equity Factor Models, July 27, 2015)

2. US Equity Stretegy: Growth Matters, May 28, 2013
- 있음

3. US Quant Research: Idiosyncratic Risk Returns, Jan. 11, 2013

4. US Equity Streategy: Seeing Industries in a New Light with PRISM, Sept. 3, 2013

5. US Equity Strategy: Cyclicals or Defensives, Feb. 26, 2012

6. US Equity Strategy: Introducing MOST : Morgan Stanley's Quantitative Stock-Selection Model, Feb 13, 2011)

7. US Equity Strategy: Introducing BEST: Morgan Stanley's New Biannual Equity Selection Tool, Seb 11, 2011)

8. US Equity Strategy: Unemotional Approaches, Oct 4, 2011

9. US Equity Strategy: The Factor Reference Guide, Aug 5, 2014

10. US Quant Research: Quantabees Beware: Explaining Factor Efficacy, Mar 7 2012

11. US Quant Research: Making Factor Consistency Pay Off, Dec 7 2015

12. Global Quantitative Research: Finding Alpha in Global Equity Markets Sep 21 2015

12. US Quant Research: Do Moving Average Crossover Signals Work in Equities? April7. 2016

<Nomura>
Mezrich, Ishikawa, “Decision rules to forecast momentum.” Nomura, 4 September 2012.

20160611(토) - Risk off?

<Facts>
1. Dax : -2.87%
2. S&P E-mini : -0.85%
3. The grobal railly in government bonds
- German, UK and Japanese sovereign bond yields all reached historic lows
- 10Year Bund : 0.01%
- 10Year JGB : -0.15%
- 10Year US-T : down 4bp at 1.64% ( 2Year : 0.73%)
4. Brent oil : -2.7% ( but still up 1.8% on the week)

<Why?>
1. an anxiety over the world economy
- real fears in the markets about global growth
- watch out the price of copper!
2. Britain's referendum on EU membership sent investors racing to safety.
- The polls on Jun 23
- "UK political risk has become the biggest deal in global financial markets, for now" - City Strategist Jonathan Stubbs
3. do not know the effects of negative interest rates in the eurozone and Japan
- Bill Gross : "Supernova that will explode one day"

<Next week>
1. 6.16(Thursday)
 - FOMC & BOJ