Demystify systematic macro?

Would anyone care to explain simply:

1) Why do people say that global macro is moving away from broad-based discretionary macro, of the likes of Soros and Druckenmiller, into increasingly systematic and quant macro

2) How does it actually work, from a bird's-eye view? Because the way I envision it, macro problems are very broad, general and assumptive. Reducing these problems into quant problem sets would be a bit challenging? And even after solving some quant problems, it would still need to go thru more chains of assumptions before it manifests in a solution at the portfolio level

3) What sort of work is done at the junior level. What are they solving and modelling?

4) Any good resources on systematic macro?

 
Most Helpful

I studied quite a few papers during my masters. 2 that seem relevant to this question are:

  • Time Series Momentum (TSMOM) – Moskowitz, Ooi, Pedersen (MOP)
  • Demystifying Managed Futures – Hurst, Ooi, Pedersen (HOP)

The HOP paper is an extension of MOP – it's key finding is that TSMOM (time series momentum) is the main driver of managed futures returns (systematic macro, or CTAs)

Here are my summary notes for those two papers:

MOP

  • A diversified portfolio of time series momentum strategies across all asset classes

    delivers substantial abnormal returns

  • Little exposure to standard asset pricing factors and performs best during extreme

    markets

  • speculators profit from time series momentum at the expense of hedgers: speculators

    trade in the same direction as a return shock and reduce their positions as the shock

    dissipates, whereas hedgers take the opposite side of these trades

  • paper includes country equity indexes, currencies, commodities, and sovereign bonds over more than

    25 years of data

  • past 12-month excess return of each instrument is a positive predictor of its future

    return

  • time series momentum or ‘‘trend’’ effect persists for about a year and then partially

    reverses over longer horizons – too far past fundamental = end of trend

  • findings are robust across a number of subsamples, look-back periods, and holding

    periods

  • Rather than focus on the relative returns of securities in the cross-section, time series

    momentum focuses purely on a security’s own past return

  • correlations of time series momentum strategies across asset classes are larger than the

    correlations of the asset classes themselves: stronger common component to time

    series momentum across different assets than is present among the assets themselves

  • driving force: significant positive auto-covariance between a security’s excess return

    next month and it’s lagged 1-year return. (covar between itself at different times)

  • TSMOM challenges the random walk hypothesis. Not a compensation for risk. Performs

    best in extreme market moves

  • Size each position (long or short) so that it has an ex ante annualized volatility of 40%

  • the markets we study vary widely in terms of the type of investors, yet the pattern of

    returns remains remarkably consistent across these markets and is highly correlated across very different asset classes

HOP

Extension of MOP – TSMOM is the main driver of managed futures returns

  • potential sources of profit due to initial under-reaction and delayed over-reaction to

    news

  • time series momentum strategies produce large correlations and high Rsquares with

    Managed Futures indices and individual manager returns, including the largest and most

    successful managers.

  • While the largest Managed Futures managers have realized significant alphas to

    traditional long-only benchmarks, controlling for time series momentum strategies

    drives their alphas to zero

  • Fung and Hsieh (2001) find that portfolios of look-back straddles have explanatory

    power for Managed Futures returns, but these look-back straddles are not

    implementable as they use data from future time periods

  • Finds strong evidence of trends across different look-back horizons and asset classes.

  • A time series momentum strategy diversified across all assets and trend horizons

    realizes a gross Sharpe ratio of 1.8 with little correlation to traditional asset classes.

  • in prolonged bear markets, time series momentum takes short positions as markets

    begin to decline and thus profits as markets continue to fall

  • Regress Managed Futures indices and manager returns on time series momentum

    returns, we find large R-squares and very significant loadings on time series momentum

    at each trend horizon and in each asset class.

  • In addition to explaining the time-variation of Managed Futures returns, time series

    momentum also explains the average excess return.

  • Controlling for time series momentum drives the alphas of most managers and indices

    below zero. The negative alphas relative to the hypothetical time series momentum

    strategies show the importance of fees and transaction costs.

  • strategy goes long if the preceding 1-month excess return was positive, and short if it

    was negative. The 3-month and 12-month strategies are constructed analogously. Each

    strategy always holds a long or a short position in each of 58 markets.

  • Constant volatility position sizing (MOP):

    1. Diversified portfolio not overly dependent on the riskier assets (important given large dispersion in vol for different assets: nat gas vs bonds)

    2. Risk of each asset stable over time

    3. Minimises risk of data mining

They scale positions so that each asset has the same ex ante volatility at each time

  • means that, the higher the volatility of an asset, the smaller a position it has in the portfolio, creating a stable and risk-balanced portfolio

I can dig out some more later if this is useful!

 

Systematic macro has two forms: trend following and relative value

Trend following in its basic form looks at the correlations between economic data and asset prices, then creates a rules based system to take advantage of those correlations.

Relative value in its basic form looks at the value of an asset class relative to other asset classes and profits from mispricings among asset classes. In other words, it tries to take advantage of mean reversion across assets.

 

Nothing really to de-mystify... It's mostly CTAs using trend following algos to trade listed futures... FWIW people aren't 'moving away' from discretionary macro, it's just streaky (and always has been).

Relative value is probably what you're thinking of, but that's completely different from "systematic and quant macro" and is still almost entirely discretionary. 

 

Agreed, I consider that in 2000s the banks all had prop trading groups with many FICC products. Back then you could just size up for days as a customer came in or so, swing big.

Today those FICC strategies exist and mostly at MMHF where you have to actually care about your sharpe and so on. Systematic just makes sound much safer to investors and cooler to the street.

 

Very high level zero analysis. Soros and Druckenmiller would hit an asset hard, diversification be damned. If you’re right, you’re rich. Now things are so focused on the computer it is a different game entirely. Think about trade routing in the 90’s vs today.

Only two sources I trust, Glenn Beck and singing woodland creatures.
 

why do people pay for macro research? It seems so complex with so many variables.  

 

1) Just IMO, it’s hard to have a high sharpe trading discretionary. In the end, sharpe is correlated with number of bets, and with systematic it’s easier to have a larger number of bets. You can still do it (e.g. Haidar Capital Management), but I don’t think there’s that much institutional demand for it.

2) Momentum, carry, risk premia, seasonality etc… You don’t have to just trade on super macroeconomic trends, although people do. There are more tangible opportunities when you look at individual instruments or portfolios of related instruments.

3) People generally are more quantitative and research focused. Programming, setting up models (as in statistical models rather then bottom up equities models) and testing hypothesis against past data.

 

any thoughts on where to start a career in macro today. Seems like the future of macro funds is in flux. Alan Howard went hard on Italian bonds and made big money for Brevan last year…had a small renaissance for macro. Wil they die out if we return to a macro regime similar to 2012-2019?

If you can’t get a role out of undergrad, where’s the best place to start? FICC S&T seems unappealing as S&T seems doomed.

 

any thoughts on where to start a career in macro today. Seems like the future of macro funds is in flux. Alan Howard went hard on Italian bonds and made big money for Brevan last year…had a small renaissance for macro. Wil they die out if we return to a macro regime similar to 2012-2019?

If you can’t get a role out of undergrad, where’s the best place to start? FICC S&T seems unappealing as S&T seems doomed.

 

What I can say is that few people ever expect to land a top macro role out of undergrad. Those seats are way too rare for that sort of wishful thinking. I don't think FICC trading is that bad of a start - plenty of traders from FICC desks move to global macro

 

yeah I just saw Brevan, BlueCrest, Bridgewater, Element, Tudor, Citadel GFI, DE Shaw Oculus, Graham Capital, multi managers like Exodus, Millenium, Balyasny, and related fixed income funds like Bracebridge and Capula and AQR global trading all had analyst programs on my schools handshake. I'm assuming since the S&T pipeline is drying up they are trying to attract top talent away from IB which doesn't prepare somebody to work in macro?But still your right that there is probably only 1-2 seat at all these places except Bridgewater, so wishful thinking.Do you know if FICC strategist roles have good macro exits?

 

Haidar doesn't have a high sharpe.. definitely sub 1

Yup, my bad on the confusing wording. I meant some people can still run a more volatile/low sharpe mainly discretionary Macro book, it’s just less common these days. 

 
trying_my_best

Would anyone care to explain simply:

1) Why do people say that global macro is moving away from broad-based discretionary macro, of the likes of Soros and Druckenmiller, into increasingly systematic and quant macro

2) How does it actually work, from a bird's-eye view? Because the way I envision it, macro problems are very broad, general and assumptive. Reducing these problems into quant problem sets would be a bit challenging? And even after solving some quant problems, it would still need to go thru more chains of assumptions before it manifests in a solution at the portfolio level

3) What sort of work is done at the junior level. What are they solving and modelling?

4) Any good resources on systematic macro?

1) Like others have touched on, the days where the founders go balls deep based on their macro thesis are long gone. For example the last few years Brevan has moved away from Alan running big chunk of the risk to a more diversified fund approach. It’s all about generating returns within tight risk parameters these days.

2) You don’t have to model everything. Maybe in your mind you’re thinking to do systematic macro one has to build a multivariate model including unemployment, inflation, SLOOS and God knows what, but it doesn’t have to be that way. At the end its all about finding predictive features, nothing more nothing less. And systematic macro is much much more than just trend or carry or seasonality. It doesn’t take much brain power to realise that things like term structure, vol surface etc have some predictive power - assuming you know what you’re doing ofc.

3) Mostly doing anything that your PM asks you to do. It can range from forecasting macro data, to coming up with a strategy based on the PM’s thesis/observations etc.

4) Can’t remember them on top of my head.

 

Quae ut sequi expedita perspiciatis qui praesentium ratione. Ratione odio consequatur est amet et explicabo consequatur veniam. Quis corporis enim quo et voluptas voluptatem.

Career Advancement Opportunities

May 2024 Hedge Fund

  • Point72 98.9%
  • D.E. Shaw 97.9%
  • Citadel Investment Group 96.8%
  • Magnetar Capital 95.8%
  • AQR Capital Management 94.7%

Overall Employee Satisfaction

May 2024 Hedge Fund

  • Magnetar Capital 98.9%
  • D.E. Shaw 97.8%
  • Blackstone Group 96.8%
  • Two Sigma Investments 95.7%
  • Citadel Investment Group 94.6%

Professional Growth Opportunities

May 2024 Hedge Fund

  • AQR Capital Management 99.0%
  • Point72 97.9%
  • D.E. Shaw 96.9%
  • Magnetar Capital 95.8%
  • Citadel Investment Group 94.8%

Total Avg Compensation

May 2024 Hedge Fund

  • Portfolio Manager (9) $1,648
  • Vice President (23) $474
  • Director/MD (12) $423
  • NA (6) $322
  • 3rd+ Year Associate (24) $287
  • Manager (4) $282
  • Engineer/Quant (71) $274
  • 2nd Year Associate (30) $251
  • 1st Year Associate (73) $190
  • Analysts (225) $179
  • Intern/Summer Associate (23) $131
  • Junior Trader (5) $102
  • Intern/Summer Analyst (251) $85
notes
16 IB Interviews Notes

“... there’s no excuse to not take advantage of the resources out there available to you. Best value for your $ are the...”

Leaderboard

1
redever's picture
redever
99.2
2
Secyh62's picture
Secyh62
99.0
3
Betsy Massar's picture
Betsy Massar
99.0
4
BankonBanking's picture
BankonBanking
99.0
5
dosk17's picture
dosk17
98.9
6
kanon's picture
kanon
98.9
7
GameTheory's picture
GameTheory
98.9
8
CompBanker's picture
CompBanker
98.9
9
bolo up's picture
bolo up
98.8
10
Linda Abraham's picture
Linda Abraham
98.8
success
From 10 rejections to 1 dream investment banking internship

“... I believe it was the single biggest reason why I ended up with an offer...”