The Discovering Markets Hypothesis - Worth a close look to add to our thinking of market dynamics

There are alternative views on how markets operate that are different from the efficient markets hypothesis which has taken a beating since the development in behavioral finance. Andrew Lo came up with the concept of Adaptive markets, but Tom Mayer and Marius Kleinheyer have developed an alternative called the Discovering Markets Hypothesis (DMH) which is based on three key points. Information held or learned by investors should be viewed as subjective not objective knowledge and this knowledge is adjusted when it compared against the behavior of others. Investors will communicate with others to cross-check their subjective knowledge. The communication of knowledge is through narrative. Narratives compete through their influence on prices. 

Facts influence subjective knowledge which is then shared with others through narratives. These narratives compete with others through their influence on prices. Prices, of course, will then provide feedback on the quality of the narrative. As subjective knowledge changes, there will be an impact on prices. Facts or new information will drive the changes in subjective knowledge. Because subjective knowledge cannot be counted or measured with certainty, there will be inherent uncertainty in markets which will cause prices to change in ways that are not always expected. 

The wild card for bond yields - the term premium

 

Treasury yield have moved higher while the Fed has lowered interest rates. This was not supped to happen. The question now is determining what is the fair value for Treasury yield out the curve. The usual method is to determine the real rate of interest which usually is about 2% plus some estimate of expected inflation which can be argued to be at level above 2%. So, if we use a real rate of 2% and an inflation rate of 2.5% we are at 4.5% as a good starting point; however, we need to add a term premium for the risk from holding these bonds out the curve. That number has been negative for an extended period but is now positive and rising. The current term premium is at the highest level in over a year; however, a longer history suggests that it can increase by a multiple of the current levels. One reason could be the lower liquidity in Treasuries, yet the recent fall is not seen in the term premium. Forecasting the term premium is now the Treasury yield forecast will card. 

Trend-following with industry groups - It works

 

In the paper, "A Century of Profitable Industry Trends" the authors explore long-only trend following for a 48 industry portfolio for just under a one hundred year period. They find that the simple trend timing strategy will lead to higher returns, lower volatility, higher Sharpe ratio and less downside risk. This strategy is easy to implement and is shown to work with sector ETFs with a smaller portfolio of 31 industries. The numbers are compelling and again show that a trend strategy can be effective. The strategy is effect with equal weighting, timing and sizing allocations. It may not work during every period, but the long run return beat the simple strategy of holding the market portfolio.

There is no equity risk premium!

There are several ways of measuring the equity risk premium. The above chart is a simple approach, and it shows the premium is at levels not seen since the dot-com bubble. Other approaches may show the premium slightly higher, but all are signaling that we are are at low levels which suggests that forward returns are likely to be lower.  

This is not saying returns will be negative in the next year nor does it say we will have a crash. It is saying that your long-term allocation to stocks is unlikely to generate returns similar to what were present over the post-COVID period. 

Replicating trend-following beta is achievable

 

Trend-following strategies have proved to add significant value to a portfolio, but there is a problem - what model or managers should I use to represent my trend exposure? This issue is discussed in the paper, "In pursuit of trend-following beta: The promise and pitfalls of replication".

Building your own trend model will have problem because here is a wide dispersion in the performance of different trend strategies. How do you pick the best strategy. However, there also is a problem if you want to pick specific managers.  There is wide dispersion across managers. The alternative is to hold an index or attempt to replicate a well-defined index like the BTOP50; however, any replication scheme has issues. One choice is to form a portfolio of managers, but there are ongoing costs with this fund of funds structure. An investor may be able to cut costs with forming a portfolio strategy replication; however, there may be skill leakage. You will not be able to capitalize on the stronger behavior of better managers.

The authors conduct a useful exercise on how to develop a replication strategy using a limited number of markets and simple OLS regression or LASSO regression. While not perfect, the cost savings with careful regression work can add value relative to a trend-following benchmark (more precisely a portfolio of CTA that are mainly trend-followers). My experience suggests that replication is not always easy. There are operational costs that can be high if there is not scale, but the idea of forming a cheaper replication of a manager bundle is useful and can add serve as an alternative to picking a bundle of managers.

VIX and corporate bonds

 

There is risk in corporate bonds and this risk can be measured through the VIX market. Equity and bond markets are linked. This should not be surprising since the corporate structure can be divided in equity - the residual value of the firm which is call option and bonds which are a short put option on the firm value. A change in volatility will lead to a change in the value of these options. If the VIX serves as a proxy for equity volatility, then there should be a link with bond spreads which are the added risk above Treasuries for holding corporate bonds.

The authors of an early draft paper, "The VIX as Stochastic Volatility for Corporate Bonds" show that adding the VIX to a time series model of corporate spreads will improve the time series model. First, the residuals of corporate bonds spreads are not Gaussian white noise; however, if you scale the residuals by the VIX to standardize, you will get an improved model. The spikes in spread are dampened when they are scaled by equity market volatility. This is a simple model and test, but it serves as an important improvement over looking at just the simple time series of spreads.

The economic of information - a history which leads to narrative information

 

The economics of information have made many advancements since the theory of market efficiency. The current direction of information economics is to link narrative economics with market sentiment and returns. This chart is from Ronnie Sadka of Boston College from his work with State Street on market narratives. There are several media-driven reservoirs that can be used to generate narrative information. There is information intensity as well as news sentiment. Once we have isolated information, we can look at dispersion, disagreement, and content. From the narrative information portfolios can be formed with narrative sensitivity. 

Ways of thinking about news narrative that can be quantified.

Trend-following with stops reduces drawdowns

 

A core question with any trading strategy is whether or not to use stops. The answer is not always clear. It is clear that having stops too close will reduce return because  many good positions are lost to reduce risk, it is often an empirical question of the value of this risk management tool. A recent paper set-up a simple model across all major assets classes to determine the value of stops. This work is not definitive and is just an example, but it does add to the discussion.  See "Cross-asset trend following algorithm".

I will not go through all of the assumptions to get these results, but the base model represents a long-term trend model based on a moving crossover model using 50 and 100 days with stops based on a multiplier using the average true range. The data includes 39 assets over a 32-year period.

For the long/short portfolio there is a slightly lower Sharpe ratio when stops are used, but there is a significant decline in the maximum drawdown. A stop-loss will not support higher returns. A review of the cumulative returns chart shows a marked difference in the return pattern.

Risk cycles and the business cycle - Form low risk comes growth but also greater credit risk

There is a business cycle, a credit cycle, and a risk cycle. These are all connected. In the paper, "The impact of risk cycles on business cycles; a historical view" the identification of the risk cycle is used to generate statement on growth, risk-taking, and the potential for crises. 

The authors show that a perceived low risk environment will encourage risk-taking which will support stronger growth, but with this growth comes greater financial vulnerabilities, that is, there will be a reach for more risk and for higher leverage. The duration of low-risk impacts growth.

Increasing financial vulnerability will lead to a reversal of growth when these risks are realized. From a low-risk cycle comes a growth upturn which creates excess credit growth. If this happens on a. global the impact on growth will be stronger because there will be grater movement in global capital flows. 

This is another way of saying that Minsky behavior can be measured through a risk cycle.

VIX and the equity markets - equities drive volatility

 

We know that there is a correlation between the VIX and SPX. A spike in volatility will be associated with lower equity returns, but the next question has not been fully explored. What is the direction of causality? A recent paper looks at this specific question, see "Which way does the wind blow between SPX futures and VIX futures?". The authors look at Granger causality tests with the method of identification through heteroskedasticity and find that the SPX futures is causally prior to the move in VIX futures. There will be a shock in new information that impacts the equity futures which then moves to the volatility futures. The authors do a good job of looking more closely at the data across some key time periods to find the correct causal relationship. 

Trend-following in high and low rate regimes

We know that trend-following can add to a portfolio mix of equity and bonds even with bonds being a diversifier; however, there is a question on whether trend-following will add value when rates are high. We know that bond volatility will be higher when rates are higher, but if you can get a higher rate, it provides a hurdle that the trend manager must overcome. For the trend-following to work it must be better than the bond rate and bond returns. The folks at Quantica Capital have another take on this issue in their piece "The Additional Benefits of Trend-following when rates are high". They find that at high rates, the correlation between stocks and bonds increase. If that is the case, there will be a good reason to give a higher allocation to the trend-following diversifier. 

The study focuses on three charts. The first chart sets the stage through showing the return, risk and correlation for stocks and bonds and shows why you should diversify over these two assets. The next figure shows trend-following return and risk in low and high-rate environments as well as the correlation between stocks, bonds and trend-following. Note that trend-following is more volatile in high-rate environments. It is also has slightly lower returns, but the correlation is very different. Trend-following is less correlated to stocks and bonds in high-rate environments, so there is greater benefit for trend-following in a portfolio. 

If rates are above average, then a simple optimizer will show that you should have on average an increase to trend-following of almost 3x versus a low-rate environment. This is easy to do and should be considered by many allocators.

Using our experiences is not always good

 

Follow our experience because as we gain more experience, we become better at making decision. Our wisdom comes from our experience. The Myth of Experience: Why we learn the wrong lessons and ways to correct them by Emre Soyer and Robin Hogarth is another take on behavioral mistakes and the problems of psychology on our decision-making. Their conclusion is that we often take-away the wrong conclusions from our experiences. Experiences that are no assess and filtered will give you the wrong answers. More experiences with the wrong assessment will make you a worse decision-maker. we use experience through linking our actions with results, but if there is not close link between the two, we will find a connection that is often wrong. The authors start with a great example. Learned people used bloodletting for centuries because they thought it worked. You bleed as a cure for a sickness and survive. It must have been the bloodletting that worked. 

We often forget or don't think about what is missing from our experiences. We do not account for the irrelevant. 

Robin Hogarth recently died, and this was one of his last books. He was one of the great researchers on decision-making and human behavior. Tversky and Kahneman have received most of the attention in this area, but Hogarth was a critical researcher in this area one the last 50 years.

Volatility is a driver for financial crises (Minsky low volatility)

Volatility is a key driver and indicator for financial crises. This volatility prediction is not what you may expect. It is known that during a financial crisis volatility will surge higher, but what is critical for determining whether there will be a crisis is the past volatility. 

What has been found is that a period of low volatility or calm markets will lead to future financial disruptions. This can be viewed as a verification of the Minsky instability hypothesis. See "Learning from History: Volatility and Financial Crises".

You could call this the "volatility paradox", low volatility will increase the chance of systemic event.  If there is prolonged low volatility, there will a higher likelihood of a banking crisis. Form a low volatility regime, there will be excessive credit build-ups and higher balance sheet leverage. You feel like there is less risk and you will then take on more leverage. This work finds that "stability is destabilizing". 

Given the long history studied and the long lag periods, it is hard to use low volatility as a trading signal for short-term shocks, but this volatility relationship is important when thinking about long-term crisis risks. Low volatility will cause investors to take bigger risks. The costs of these risks will have to be borne by someone. 

Using the TIMEMIXER approach for volatility forecasting

An application of time mixing for volatility forecasting can be an important advancement for risk management. Research has extended the work on GARCH to an extreme, but there may be other techniques in time series forecasting that can be applied to financial time series that may be very useful. A recent paper focused on TimeMixers which employs different time scales as a method to improve forecasts. See "Volatility Forecasting in Global Financial Markets Using TimeMixer".

The idea behind TimeMixers is straight-forward. There is imbedded in any times series relationship with different timeframes that can exploited. There can be long-term seasonality. There can be cycles or trends that are longer than a few days that will not be captured with daily data. Classic time series in ARMA models can handle seasonality and can identify autocorrelation at different lengths, but a more explicit breakdown of data may improve forecasts. 

I like the technique used and the author applied it to a broad set of markets, but I was disappointed that there was no testing against other types of models for volatility. This process looks interesting but it is not clear it is any better than what we already have. The MAE, MSE, and RMSE all are low especially for short-term forecasts, but the quality of technique must be balanced with the results, the ease of understanding, and the ease of implementation. This paper does not make that strong relative case for TimeMixer ML. 

That sinking feeling in the CMBS market

 

Sometimes you get hit with a wave, but often you drown because the water level just gets too high and you don't have the energy to find the shore, or worse, there is no shore. There is a problem in the CMBS problem with some very good properties underwater with the sinking still occurring. We can start with the issue of running a CMBS on a single property. There is no diversification. Investors may expect that lower rated tranches will have more credit risk, but in these cases, we are seeing risk with the A tranche. Ouch. This is what is supposed to be avoiding with structuring. A combination of poor engineering with a bad environment spells disaster for investors. 

Can there be money to be made? Yes, however, it is not from dropping these bonds in a portfolio and cutting some coupons. This is risky bond trading. 

TimeMixers as a new time series tool

 

Many trend-followers are time mixers. What has been old for some is now something new. A new technique in machine learning time series forecasting that is being given more attention is a concept called time mixing. The idea of time mixing is simple. Decompose any time series into a set of different time scales. For example, there can be short-term behavior, longer-term, seasonal and very long-term behavior in prices. This breakdown is certainly true for almost all commodities. If we can take the multi-scale data, we can break the problem into two parts or blocks:  the past-decomposable-Mixing block PDM where the information at different scales is learned by the model. From the PDM, the model is sent to the Future-Multipredictor-MIxing (FMM) block which will combine the scaled information into a forecast. This type of model may not do well for very short-term predictions, but longer-term forecasts can be improved by this technique. Given this is a ML technique, the value is found in improved forecasting. This is not a structural model of how the market behaves.

What do many trend-followers do? They will often use trends for different scales or look-backs to blend into a single signal. The weighting of these signals can be a very simple, like an equal-weighted scale, or it can be optimized based on preferences or on past forecasting success. This is not the sample as a formal time mixer model, but the idea behind this approach is the same. Use different time scales to capture the different behaviors in asset time series. 

Causal structures for equity risk factors evolve over time

 

There has been a strong increase in the study of causal inference in finance. Given the increase in the factor zoo and p-hacking, there is a need to go back to basics and think about causality and what it takes to find causal relationships. 

The paper, "The Evolving Causal Structure of Equity Risk Factors" focuses on using causal structure learning methods to analyze 11 risk factors in the US equity markets. The relationships have become more sparse with the market beta having a strong influence on other risk factors. Much of this research work starts with structural vector autoregressive models which look at the impact past returns on current returns to measure the impact of cross-factor causation. SVAR models can be restructured as directed acyclic graphs (DAG) that can provide useful insights on causality. Inferring causality from these insights is the objective of causal structure learning. 

The number of edges for both correlation and causal networks have fallen through time. The markets are more sparse and more focused on instantaneous relationships. The authors also find that the number of edges in the network is negatively related to VIX shocks; however, the number of edges in the causal network will increase with a volatility shock. The volatility shock is a large increase (above the 95th percentile) in VIX between the close and open values standardized. 

This work is just a start. What is more important is finding casual relationships that can be exploited for profit.

The evolution of the hedge fund industry - The development of multi-strats

 

The hedge fund industry has gone through significant changes over the last 20 years with a recent paper providing a good description of how hedge fund structures have adapted especially since the GFC. See from Clear Alpha Technologies "The Evolution of Alpha: Exploring the past, present, and future of investing in alpha". I don't like the title because this is not about alpha creation but how an industry has adapted to changing tastes by the market to deliver what is expected to be a better alternative investment product. The alpha creation from the bundling of strategies is different from how returns are delivered to investors which is the topic of this paper.

The hedge fund industry started with the development of specialized money management funds which usually had a focused objective. From traditional long-only managers came the broader focus of long/short managers. Unfortunately, there are many ways to generate return and alpha, so investor demand bundled products in the form of fund of funds. Unfortunately, many fund of funds did not delivery on the expected promise of higher returns and controlled risk. 

Fees were higher relative to what was delivered so the market looked form other alternatives. One approach was to bring the investing in-house as market knowledge of hedge funds increased. The suppliers of returns developed in a different direction through creating multi-strat programs under a single structure. This was extremely successful, but there were limited in the ability of these multi-strats to gather the investment talent relative to demand. A new development was the multi-strategy, multi-manager fund which blended internal and external managers in a single structure. In both cases, the multi-strategy approach allowed for better risk oversight and more effective management of capital allocation. 

The other development from the multi-strategy structure was the pass-through of costs. This pass-through model protected the hedge fund business and had the ability to provide a better cost structure to investors if the costs could be controlled. Investors were now investing in hedge fund businesses not giving money to be managers. There is a nuanced difference but has a significant impact on both the suppliers and demanders of alpha. The suppliers can invest in technology and risk management and reduce fluctuations in revenue. Investors, the demanders of hedge fund services, can form strategic partnerships but may have to pay a premium for the services of these multi-strat managers. 

Hedge fund management is a very dynamic industry which will evolve as alpha changes, and the management structure will impact alpha. 

Dynamic Factor Allocation with Regime Switchng

When should you change the allocation of your factor exposure? This is a critical question and ultimately it is a question of market timing. But how do you effectively market time? This timing can be done through looking for regime switches. If we can identify if we are in a bull or bear regime, we should be able to adjust the allocation across factor allocations. Of course, if we could be perfect timers, we would not think about factor exposures, we would just go long the risky asset in a bull market regime and go short in a bear market. Alternatively, we focus on building a portfolio because we know that we will not be perfect in our timing ability.

In the new paper, "Dynamic Factor Allocation Leveraging Regime-Switching Signals", the authors look at seven long-only indices which include the market portfolio and value, size, momentum, quality, low volatility and growth and employ a Black-Litterman model with regime switching. They show that the dynamic portfolio will do better than an equal-weighted portfolio when the allocation decisions are based on a sparse jump (SJM) model. The SJM uses risk and return measures to identify the regimes. It is a form of cluster analysis to find bull and bear behavior across the set of factors analyzed. This regime model is the novel addition to this paper. A regime model is applied to each factor, so each has identified their own bull and bear period.

If you can identify regimes correctly, you can market time. That should seem obvious, and it is. There are many ways to form regimes, so the key job of the PM is to find the right way to identify these regimes.

Note that different regime modeling techniques will generate different regimes for the same data. The regimes for the size factors are very different.

Whether this can be implemented in world of transaction costs, is questionable, but the key concept of using regimes to help build portfolios is very useful.