Viewing entries in
Quant

Comment

The Greenspan Put is coming

With markets in turmoil, inflation all but non-existent, and global GDP growth stalling, the FED’s next significant move will be to follow many other central banks and ease. As interest rates have only moved up by a quarter percentage it is likely quantitative easing or negative interest rates are coming. Most major market participants are now aware of this, opinions on whether it is a good idea are divided. At Davos, some attendees were calling for global central banks to stop easy money policies and let economies correct. In this note we explain why this is unlikely and how we will take advantage of the FED’s next move.

Household net worth, debt, business spending, and QE

One of the main purposes of quantitative easing was to inflate asset prices. The idea is: when asset prices increase consumers feel confident and start spending, businesses hire and invest, inflation rises, and the economy normalizes. Our analysis shows only moderate correlation between household net worth and future consumer spending (Chart 1). The correlation between net worth and consumer spending peaks out at ~0.60, with consumption lagging net worth by two months. One of the reasons for the moderate, not strong, relationship this is that financial assets comprise an ever increasing share of net worth. As most financial instruments are marked to market daily, changes in market price cause wild swings in consumer and business sentiment (Chart 1 and Chart 2).

Chart 1: Change in net worth and household spending 

Source: FRED, Vital Data Science Inc. 

Source: FRED, Vital Data Science Inc. 

Chart 2: Composition of net worth

Source: FRED, Vital Data Science Inc. 

Source: FRED, Vital Data Science Inc. 

As we’ve pointed out in a previous note, many American consumers have been left behind by an economy with global supply chains, knowledge sector growth, and the rise of low wage and temporary work. Instead of the intended purpose, easy money policies have artificially inflated asset prices. Haunted by the great recession and left behind by the economy, consumers have focused on deleveraging and saving instead of spending, which is one of the reasons why inflation has not taken off. As consumers deleverage businesses have taken on debt, and instead of using the debt to invest in the future they have used it to artificially inflate earnings per share through share buy-backs (Chart 3). Large share buy backs mask the fact that earnings and sales in S&P 500 firms have declined throughout 2015, pointing to weakness in the global economy. Some market participants are beginning to tire of the financial engineering and beginning to push for change - one reason why short term focused activist funds are starting to fall out of favour.

Chart 3: S&P 500 share buy backs 

Source: FactSet, http://www.factset.com/insight/2015/12/buybacks_12.15.15#.Vr9xC8eMD-Y

Source: FactSet, http://www.factset.com/insight/2015/12/buybacks_12.15.15#.Vr9xC8eMD-Y

Internationally, corporate and government debt binges have fuelled over capacity in China which is now experiencing slowing growth as a result. As China was the main driver of global GDP growth out of the great recession (Chart 4), and a large consumer of commodities (Chart 5a and 5b), we can expect the global economy to slow.

Chart 4: World and China GDP 

Source: World Bank, Vital Data Science Inc.

Source: World Bank, Vital Data Science Inc.

Chart 5a and 5b: World and China Imports/Exports

Source: World Bank, Vital Data Science Inc. 

Source: World Bank, Vital Data Science Inc. 

In short: today’s macro economic climate does not support current asset valuations. Investors are irrationally buying assets at high valuations because they believe that easy money policies support high asset prices and/or the macro economic climate will eventually catch up to the market. Central banks are easing monetary policy in an attempt to push economies to catch up with asset valuations and keep asset valuations high. Since consumers are deleveraging, monetary easing is having a limited impact on the real economy.

What happens next

At Vital Data Science, we analyze the global economy and asset prices several ways. One process we go through is to collect hundreds of thousands economic indicators, reduce and extract features, then summarize the global economy in a three dimensional visualization (Figure 1). While this tool over-simplifies the complexities and interactions in the global economy, it is useful for spotting long term trends. Our representation shows the economy peaked early 2015 and has been in decline since. We extracted features from this representation and built a model using the features as variables. We found markets lag our bull/bear indicators by 12-18 months. Back testing showed the model was ~70% accurate in predicting market swings. This model is currently predicting a bear market.

Figure 1: Three dimensional representation of the the global economy 

Source: FRED, Vital Data Science Inc. 

Source: FRED, Vital Data Science Inc. 

Another approach we take is to run an algorithm through the macro economic statistics we collect and develop indexes representing areas of interest in the global economy. We train a model on the indexes and forecast market movements. Our model, when back tested 20 years, is 87% accurate in predicting major market swings. This model is also predicting a bear market.

We believe that current distortions in the global economy have deviated from historical patterns in two key ways:

  • Increased interdependence in global economies through global supply chains; and
  • worldwide acceptance of unorthodox central bank policies like quantitative easing and negative interest rates.

Models based on historical information cannot accurately reflect these new dynamics without a significant adjustment of forecasted results. Taking all of this into account, our current base case global macro economic scenario is:

  1. The US is not immune to the global slow down and the heat experienced at the tail end of 2015 will cool by the end of the first half of 2016.
  2. The Greenspan put is coming. You can call it the Yellen put. The FED will not raise interest rates, it will lower them, and then return to quantitive easing or negative interest rates by 2017.
  3. US equity markets will swing wildly but trend flat to negative in 2016, before the end of the year a bear market will be evident.
  4. Consumers will continue to deleverage and save through 2017. Normal spending growth will return when debt burden returns to normal levels. Then the next cycle will begin.
  5. Investors and corporate management teams will begin to adjust to the deflationary deleveraging phase in the global economy and financial engineering will peak in 2016, then begin decreasing in 2017.

The world is becoming increasingly turbulent: conflicts in the Middle East, tension between Russian and the West, and tension between China, its neighbours, and the United States all have economic implications. Global political turbulence perpetuates the monetary policy race to zero and below, and increases the time required to work through global economic imbalances (i.e. oil supply/demand). In our view, this is the largest risk in the global economy over the long term.

Strategy

Our belief that current global economic distortions cannot be accurately reflected by existing economic models, which are trained using historical data, without significant adjustment of forecasted results, was the driver of our earlier market neutral strategy idea. Currently, we see success coming from beginning to move away from market neutral and into a net market short position. We see the buffering impact of global central banks policies having a decreasing effect in supporting asset prices going forward and believe that it will take a number of years for the real economy to catch-up to current asset valuations. 

Positions

In our note in December 2015, we highlighted three pair trade ideas. We continue to believe that these positions will yield success. To take advantage of our global macro view, we also highlight other ideas below. Table 1 summarizes our position ideas and results to date. To date, our position has yielded a small loss, in the same period the S&P 500 dropped 7.5%.

Table 1: Update on existing positions 

Source: Vital Data Science Inc., Morning Star

Source: Vital Data Science Inc., Morning Star

To take advantage of our global macro view, we believe positions in US treasuries, and a net short position on the S&P 500 will yield success going forward. An example portfolio is shown in Table 2. We will optimize this portfolio as the direction of the global economy becomes more clear, in future notes. The next important check point for us is the FED’s March meeting. 

Table 2: Vital Data Science Inc. example portfolio

Source: Vital Data Science Inc. 

Source: Vital Data Science Inc. 

A note on benchmarking

As we have developed enough positions to build a very concentrated long/short portfolio, we are establishing a benchmark to validate our ideas. Our benchmark will be 70% US equities represented by SPY, and 30% US Treasuries represented by GOVT. We believe that such a ‘do nothing’ portfolio presents most investors a reasonable alternative to active trading and balancing. It is easily replicated, realistic, and presents reasonable risk/reward balance.

On a go forward basis we will measure results relative to this benchmark. If our positions expand to include global assets or alternative investments, we will adjust the benchmark. 

Disclaimer

Any material provided in this blog is for general information use only. You should not act based solely upon the materials provided herein. Vital Data Science Inc. advises you to obtain professional advice before making investment decisions. Your use of these materials is entirely at your own risk. In no event shall Vital Data Science Inc, its officers, directors or employees be liable for any loss, costs or damages whatsoever.

  

Comment

Comment

Quantessential Views: Global Macro update, FED rate increase, and top pair trade ideas

Summary points

  • Global economic picture remains murky with negative skew
  • Current equity valuations are high therefore risk is to the downside
  • A market neutral strategy is recommended, we highlight three possibilities

Economic Climate

In our piece entitled ‘Ray Dalio is right more quantitative easing is coming’ we discussed the state of the global economy and our belief that Mr. Dalio was correct in his assertion that the FED’s next big move would be quantitative easing. Low inflation, mixed jobs data, and global market chaos prevented the FED from raising rates September through November of this year. In December, expecting future inflation, but mainly to maintain credibility, the FED increased interest rates by 0.25%. Vital Data Science’s base case view is this will continue for two to four rounds before secular forces begin to impact the FED’s decisions again. We are collecting data, calibrating models, and will have an updated view early in the New Year.

3Q15 is behind us and the theme of the earnings season was the slowdown in global economic activity, especially in developing economies. S&P 500 earnings increased at their slowest rate since the recovery began (Chart 1) yet valuations remain stubbornly high. Inflation is low, and while the US labor market shows signs of picking up, significant slack remains. With slack in the labor market, a global slow down, and a stronger US dollar, we believe there is significant risk to achieving the FED’s stated goal of 2% inflation. Further,  the PMI Composite Index (Chart 2), a reasonably reliable indicator of recession risk has been in decline all year and now sits below 50. A stronger US dollar will put further pressure on the PMI Composite Index.

In short, our current macro view is:

  • the global economic picture remains murky with risk skewed negatively;
  • US recovery remains uneven, however bright spots are developing;
  • at current valuations, risks to equity markets are skewed negatively;
  • in the short term US markets will be flat, volatile, and trends hard to forecast.

Therefore we recommend a market neutral strategy - pair trades. In this piece we highlight two pair trade ideas along with one speculative pair trade. We provide the results of an analysis Vital Data Science performed, in which we found most open, cointegrated pairs of stocks on the NYSE below. For a more complete list of open pairs on the NYSE please visit part one of Quantessential Views.

Chart 1: S&P 500 earnings real growth at lowest point since recovery (<-8%)

Normal 
 0 
 
 
 
 
 false 
 false 
 false 
 
 EN-US 
 JA 
 X-NONE 
 
  
  
  
  
  
  
  
  
  
  
 
 
  
  
  
  
  
  
  
  
  
  
  
       
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
      /* Style Definitions */
table.MsoNormalTable
	{mso-style-name:"Table Normal";
	mso-tstyle-rowband-size:0;
	mso-tstyle-colband-size:0;
	mso-style-noshow:yes;
	mso-style-priority:99;
	mso-style-parent:"";
	mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
	mso-para-margin:0cm;
	mso-para-margin-bottom:.0001pt;
	mso-pagination:widow-orphan;
	font-size:10.0pt;
	font-family:Cambria;
	mso-ascii-font-family:Cambria;
	mso-ascii-theme-font:minor-latin;
	mso-hansi-font-family:Cambria;
	mso-hansi-theme-font:minor-latin;
	mso-ansi-language:EN-US;
	mso-fareast-language:JA;}   Source: www.multpl.com

Source: www.multpl.com

Chart 2: PMI Composite Index has been in decline all year and now sits below 50

Normal 
 0 
 
 
 
 
 false 
 false 
 false 
 
 EN-US 
 JA 
 X-NONE 
 
  
  
  
  
  
  
  
  
  
  
 
 
  
  
  
  
  
  
  
  
  
  
  
       
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
      /* Style Definitions */
table.MsoNormalTable
	{mso-style-name:"Table Normal";
	mso-tstyle-rowband-size:0;
	mso-tstyle-colband-size:0;
	mso-style-noshow:yes;
	mso-style-priority:99;
	mso-style-parent:"";
	mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
	mso-para-margin:0cm;
	mso-para-margin-bottom:.0001pt;
	mso-pagination:widow-orphan;
	font-size:10.0pt;
	font-family:Cambria;
	mso-ascii-font-family:Cambria;
	mso-ascii-theme-font:minor-latin;
	mso-hansi-font-family:Cambria;
	mso-hansi-theme-font:minor-latin;
	mso-ansi-language:EN-US;
	mso-fareast-language:JA;}   Source: FRED

Source: FRED

The idea behind pair trades is to pair a long position in one stock with a short position in another stock. The two stocks are to be cointegrated, meaning that a linear combination of the stocks is integrated of order zero. Generally, the firms are in the same industry and have similar beta so a pair trade creates a hedge to the industry in which the firms operate as well as the market.

 Top Pair Trade Ideas

 1. Long 3 units of CNX, short 4 units of CNQ

Quantitative data for the trade is summarized in Table 1. CNX (Consol Energy) is a NE US coal and natural gas producer. CNQ (Canadian Natural Resources) is a Canadian oil sands and heavy oil producer. Relative price performance is show in Chart 3.

Table 1: Quantitative data for CNX/CNQ pair

Normal 
 0 
 
 
 
 
 false 
 false 
 false 
 
 EN-US 
 JA 
 X-NONE 
 
  
  
  
  
  
  
  
  
  
  
 
 
  
  
  
  
  
  
  
  
  
  
  
       
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
      /* Style Definitions */
table.MsoNormalTable
	{mso-style-name:"Table Normal";
	mso-tstyle-rowband-size:0;
	mso-tstyle-colband-size:0;
	mso-style-noshow:yes;
	mso-style-priority:99;
	mso-style-parent:"";
	mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
	mso-para-margin:0cm;
	mso-para-margin-bottom:.0001pt;
	mso-pagination:widow-orphan;
	font-size:10.0pt;
	font-family:Cambria;
	mso-ascii-font-family:Cambria;
	mso-ascii-theme-font:minor-latin;
	mso-hansi-font-family:Cambria;
	mso-hansi-theme-font:minor-latin;
	mso-ansi-language:EN-US;
	mso-fareast-language:JA;}   Source: Vital Data Science, Yahoo

Source: Vital Data Science, Yahoo

CNX stock has been hammered to the point where value investors are starting to take notice. Consol Energy produces high BTU North Appalachia coal and has been able to take market share from Appalachian producers in the commodity downturn. Increasingly they are positioning themselves as a natural gas producer in the Marcellus - the best natural gas basin in the US. Natural gas is not imported into the US, in fact it is exported to Mexico, Canada, and will soon be waterborne, helping to clear the US over supply. Marcellus well economics are competitive globally. CNX is cash flow positive and has a manageable debt position. CNX trades at 5.30 EV/EBITDA, 0.33 P/B, and 0.54 P/S. CNX is well positioned to compete in a low commodity, over supplied market, and trades at a lower valuation than weaker competitors.

Chart 3: Relative price performance of CNX and CNQ

Normal 
 0 
 
 
 
 
 false 
 false 
 false 
 
 EN-US 
 JA 
 X-NONE 
 
  
  
  
  
  
  
  
  
  
  
 
 
  
  
  
  
  
  
  
  
  
  
  
       
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
      /* Style Definitions */
table.MsoNormalTable
	{mso-style-name:"Table Normal";
	mso-tstyle-rowband-size:0;
	mso-tstyle-colband-size:0;
	mso-style-noshow:yes;
	mso-style-priority:99;
	mso-style-parent:"";
	mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
	mso-para-margin:0cm;
	mso-para-margin-bottom:.0001pt;
	mso-pagination:widow-orphan;
	font-size:10.0pt;
	font-family:Cambria;
	mso-ascii-font-family:Cambria;
	mso-ascii-theme-font:minor-latin;
	mso-hansi-font-family:Cambria;
	mso-hansi-theme-font:minor-latin;
	mso-ansi-language:EN-US;
	mso-fareast-language:JA;}   Source: Vital Data Science, Yahoo

Source: Vital Data Science, Yahoo

Because CNQ is a well-managed company it pains us to call it a short. Nonetheless we believe CNQ is overvalued relative to CNX. A business can be managed flawlessly but if cut off from markets, sells a replaceable product that is heavily discounted with high production costs, and faces ever increasing regulatory and taxation hurdles, the business will have a hard time competing. As a Canadian oil sands and heavy oil producer this is the tough position CNQ finds itself in; however on relative basis its valuation has held up. CNQ trades at 6.61 EV/EBITDA, 1.11 P/B, and 2.18 P/S.

We see this gap closing within the next year driven by bad news out of Canada.

2. Long 1 unit of PII, short 4 units of UPS

Quantitative data for the trade is summarized in Table 2. PII (Polaris) is an industry leader in the powersports industry. UPS (United Parcel Service) is an industry leader in parcel delivery and freight.  Both companies are leveraged to consumer financial health. Relative price performance is shown in Chart 4.

Table 2: Quantitative data for PII/UPS pair

Normal 
 0 
 
 
 
 
 false 
 false 
 false 
 
 EN-US 
 JA 
 X-NONE 
 
  
  
  
  
  
  
  
  
  
  
 
 
  
  
  
  
  
  
  
  
  
  
  
       
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
      /* Style Definitions */
table.MsoNormalTable
	{mso-style-name:"Table Normal";
	mso-tstyle-rowband-size:0;
	mso-tstyle-colband-size:0;
	mso-style-noshow:yes;
	mso-style-priority:99;
	mso-style-parent:"";
	mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
	mso-para-margin:0cm;
	mso-para-margin-bottom:.0001pt;
	mso-pagination:widow-orphan;
	font-size:10.0pt;
	font-family:Cambria;
	mso-ascii-font-family:Cambria;
	mso-ascii-theme-font:minor-latin;
	mso-hansi-font-family:Cambria;
	mso-hansi-theme-font:minor-latin;
	mso-ansi-language:EN-US;
	mso-fareast-language:JA;}   Source: Vital Data Science, Yahoo

Source: Vital Data Science, Yahoo

Polaris recently lowered their 2015 guidance and the stock was hammered as a result. Guidance was lowered due to lower than expected Off-Road Vehicle sales and unusually warm weather impacting snowmobile sales. Top line growth was adjusted to 4%-5% compared to the previous expected value of 10%-11%. PII now trades at 1.14 EV/EBITDA, 5.95 P/B, and 1.12 P/S.

PII share price appreciated significantly in 2013 when the company was growing at a record pace, but earnings growth and share price have since matured. Polaris offers a broad product suite that hits every segment of the powersport industry. With a healthy balance sheet, industry leading products, and a management team that is dedicated to growth, Polaris has many options to get out of the current rout. We believe the recent dip presents an opportunity to buy an industry leading, innovative company for an attractive valuation.

Chart 4: Relative price performance of PII and UPS

Normal 
 0 
 
 
 
 
 false 
 false 
 false 
 
 EN-US 
 JA 
 X-NONE 
 
  
  
  
  
  
  
  
  
  
  
 
 
  
  
  
  
  
  
  
  
  
  
  
       
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
      /* Style Definitions */
table.MsoNormalTable
	{mso-style-name:"Table Normal";
	mso-tstyle-rowband-size:0;
	mso-tstyle-colband-size:0;
	mso-style-noshow:yes;
	mso-style-priority:99;
	mso-style-parent:"";
	mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
	mso-para-margin:0cm;
	mso-para-margin-bottom:.0001pt;
	mso-pagination:widow-orphan;
	font-size:10.0pt;
	font-family:Cambria;
	mso-ascii-font-family:Cambria;
	mso-ascii-theme-font:minor-latin;
	mso-hansi-font-family:Cambria;
	mso-hansi-theme-font:minor-latin;
	mso-ansi-language:EN-US;
	mso-fareast-language:JA;}   Source: Vital Data Science, Yahoo

Source: Vital Data Science, Yahoo

UPS is a giant in parcel delivery and freight, a growing but highly competitive industry. As online shopping sales sky rocketed UPS was well positioned to reap the rewards. Coming out of the 2008 downturn UPS share price has closely tracked the broad market, and like the market UPS shares have become richly valued. UPS currently trades at 11.47 EV/EBITDA, 46.23 P/B, and 1.53 P/S.

Most recently headwinds are beginning to show:

  • Amazon announced it is investigating its own parcel delivery system
  • Peer to peer economy options are rising
  • UPS has been having trouble keeping up with increased online orders as on time delivery times fell following Cyber Monday to 91% from a 97% last year

The parcel delivery business is competitive and ripe for disruption. As traditional companies fail to meet market demand adequately, online retailers will move to non-traditional options that will take market share away from incumbents.

We see this trade closing within the next one and a half years as it will take several quarters for PII team to reposition, in the mean time non-traditional parcel delivery options will continue to multiply.

3. The speculative trade: long 1 unit of MW, short 4 units of UPS

Warning: this one is not for the faint of heart. A massive spread has developed between UPS and (MW) Men’s Wearhouse, two companies whose share are cointegrated and leveraged to consumer spending. Table 3 shows quantitative data for the trade and Chart 5 shows relative performance.

Table 3: Quantitative data for MW/UPS pair

Normal 
 0 
 
 
 
 
 false 
 false 
 false 
 
 EN-US 
 JA 
 X-NONE 
 
  
  
  
  
  
  
  
  
  
  
 
 
  
  
  
  
  
  
  
  
  
  
  
       
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
      /* Style Definitions */
table.MsoNormalTable
	{mso-style-name:"Table Normal";
	mso-tstyle-rowband-size:0;
	mso-tstyle-colband-size:0;
	mso-style-noshow:yes;
	mso-style-priority:99;
	mso-style-parent:"";
	mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
	mso-para-margin:0cm;
	mso-para-margin-bottom:.0001pt;
	mso-pagination:widow-orphan;
	font-size:10.0pt;
	font-family:Cambria;
	mso-ascii-font-family:Cambria;
	mso-ascii-theme-font:minor-latin;
	mso-hansi-font-family:Cambria;
	mso-hansi-theme-font:minor-latin;
	mso-ansi-language:EN-US;
	mso-fareast-language:JA;}   Source: Vital Data Science, Yahoo

Source: Vital Data Science, Yahoo

For those who follow equity markets Men’s Wearhouse needs no introduction. The affordable men’s fashion retailer’s shares have been slaughtered, down over 60%, since management cut earnings forecasts down by 40% in November, and then reported very weak sales at Jos. A. Bank early this month. Men’s Wearhouse acquired Jos. A. Bank in 2014 for $1.8 billion, and largely paid for the deal with debt. Any value from the acquisition has been erased from MW’s share valuation, yet the debt remains. MW is now trying to change brand perception and consumer buying patterns to give the business a sustainable long term strategy. This will take time, there will be further missteps, additional write downs, and there is risk that the strategy will not work.

While the future of MW is not clear there is potential for significant upside from the current valuation, hence why it is part of our speculative trade. The company trades at 7.05 EV/EBITDA with much of the $2.3 billion EV being comprised of $1.65 billion in debt, 0.71 P/B, and 0.19 P/S - a significant discount to its direct competitors. MW legacy brands are in good shape and the firm generates $2.3 billion in revenue without Jos. A. Bank. MW is cash flow positive and focused on reducing debt. We believe that while significant risk remains, there is potential for a phoenix to rise from the ashes.

Given the overhang on MW stock, we do not see this spread closing for at least one and a half years, likely two.

Chart 5: Relative price performance of MW and UPS

Normal 
 0 
 
 
 
 
 false 
 false 
 false 
 
 EN-US 
 JA 
 X-NONE 
 
  
  
  
  
  
  
  
  
  
  
 
 
  
  
  
  
  
  
  
  
  
  
  
       
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
      /* Style Definitions */
table.MsoNormalTable
	{mso-style-name:"Table Normal";
	mso-tstyle-rowband-size:0;
	mso-tstyle-colband-size:0;
	mso-style-noshow:yes;
	mso-style-priority:99;
	mso-style-parent:"";
	mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
	mso-para-margin:0cm;
	mso-para-margin-bottom:.0001pt;
	mso-pagination:widow-orphan;
	font-size:10.0pt;
	font-family:Cambria;
	mso-ascii-font-family:Cambria;
	mso-ascii-theme-font:minor-latin;
	mso-hansi-font-family:Cambria;
	mso-hansi-theme-font:minor-latin;
	mso-ansi-language:EN-US;
	mso-fareast-language:JA;}   Source: Vital Data Science, Yahoo

Source: Vital Data Science, Yahoo

DISCLAIMER

Any material provided in this blog is for general information use only. You should not act based solely upon the materials provided herein. Vital Data Science Inc. advises you to obtain professional advice before making investment decisions. Your use of these materials is entirely at your own risk. In no event shall Vital Data Science Inc, its officers, directors or employees be liable for any loss, costs or damages whatsoever.

Comment

Comment

Quantessential Views: Sample Data

In an upcoming post, Vital Data Science will provide a Global Macro update and our top pair trade ideas. The files included below are part of the analysis. These PDFs are intended to be samples only. The data has been trimmed to save space. If you would like the complete excel format files please contact us.

Description: Correlation and cointegration data for almost every stock on the NYSE. Sorted 1) alphabetically, 2) top 300 by correlation.

File: Cointegration data for NYSE stocks 2015-12-20

Description: Potential pair trades for NYSE listed stocks. After calculating correlation and cointegration between many of the shares traded on the NYSE, we created this list for further investigation. The list is sorted based on current spread, however the spread between each stock pair is significant. Please note that just because two stocks are mathematically cointegrated and have a statistically significant spread does not mean the spread will close. Thorough fundamental analysis is needed to assess the likelihood of the spread closing, and expected timing of such closure. For three pair picks which we believe are likely to close please see our future Quantessential Views piece. 

File: Cointegrated stock pairs with significant spreads on the NYSE

DISCLAIMER

Any material provided in this blog is for general information use only. You should not act based solely upon the materials provided herein. Vital Data Science Inc. advises you to obtain professional advice before making investment decisions. Your use of these materials is entirely at your own risk. In no event shall Vital Data Science Inc, its officers, directors or employees be liable for any loss, costs or damages whatsoever.

Comment

Comment

Economic topography and its use in economic forecasting

Big data quant techniques, combined with advances in statistical programming and visualization, have created opportunities to view the economic landscape holistically. These techniques allow analysts to quickly process, analyze, summarize, and visualize hundreds of thousands economic variables and make accurate decisions quickly. These techniques are particularly useful when developing classification economic forecasting models.

Chart 1 is a graphic summary of the economic topography with the S&P 500 (orange line) overlaid on the economic landscape from 1950 to present. Chart 1 was generated by collecting, analyzing, and summarizing over 200,000 international economic variables.

Chart 1: Economic topography, 1950 to present

Source: Author

Source: Author

The challenges of big data

Working with highly dimensional data is challenging. Trying to make sense of hundreds of thousands economic indicators is not intuitive, but dimensionality can be often be reduced. This leads to faster and more accurate decision making.

Dimensionality reduction occurs in the natural world; the perfect example is human sensory data. When deciding whether the piece of food you are consuming tastes good or bad, and whether you should consume more or not, the brain will receive data from millions of sensors, extract the necessary features, and make a binary decision: eat more or stop eating. The problem can be represented as shown in figure 1.

The remainder of the post will be spent providing the reader a brief explanation of how one can go from 200,000+ economic indicators to a go/no-go investment decision.

Figure 1: should I consume more food?

Source: Author

Source: Author

Dimensionality reduction and feature extraction

After framing your problem, deciding on what data to collect, collecting, and then processing the data, the next critical step is often dimensionality reduction. Many dimensionality reduction techniques exist. One popular linear dimensionality reduction technique is principal component analysis (“PCA”), which transforms the original data set to a new coordinate system by finding the axis on which the data has the most variance. PCA achieves this by finding the eigenvalues and eigenvectors of the covariance matrix of a data set; the axis with the largest eigenvalue is chosen as the first principal component. Then, the axis with the next highest eigenvalue, which is also orthogonal to the prior axis is chosen. The process can be repeated until the number of principal components equals the number of variables in the original data set. By dropping principal components with low eigenvalues, the key features of the data are maintained while dimensionality is reduced.

Linear dimensionality reduction is very useful, however some data sets do not lend to being summarized linearly. In this case, a nonlinear dimensionality reduction technique may be employed. Just like in the linear approach, many nonlinear dimensionality reduction techniques exist. While each approach varies, the general idea is to map points from the high dimensionality manifold to a lower dimensionality manifold, while maintaining the distances between points.

Visualization

Once dimensionality is reduced, data lends itself more easily to graphing and visualization. Visualizing data is one of the easiest ways for a quant to hypothesize relationships, which can then be mathematically developed and tested. Looking at chart 1, it is difficult to see any possible relationships between the topography of the economic landscape and the S&P 500 index, however when the data is zoomed in on (chart 2) possible relationships begin to appear.

Chart 2: Economic topography, 2000 to present

Source: Author

Source: Author

Upon closer inspection of chart 2, the inference can be made that a relationship between the economic topography and the S&P 500 index exists. One dimension of the economic topography appears to precede the peaks and valleys of the S&P 500 index. While 3D plots are a useful visualization tool for quickly viewing all of the data in one window, a simple 2D plot can also provide a lot of information by graphing each variable against the response.

Feature extraction and classification

After visualizing the data, dimensions which appear to have best chances of a relationship with the response can be isolated, then classification and model building can begin. Remember, the goal is to generate a possibility matrix to layout investment decisions like we did for eating decisions.

A couple of quick ways to help classify data and extract features include using built in functions in statistical software which split data into groups, and building tree classification models. When a tree-building algorithm learns the data, the rules are excellent starting points for writing further feature extraction and classification algorithms. To maximize flexibility and testing of the program, we find that writing our own feature extraction and classification algorithms works best. This way we can target and test specific features in the data.

Model building

After the specific features have been extracted and classified, the problem can be framed per figure 2, below, which should be familiar to the reader (see figure 1). If there are ‘n’ number of features, with ‘m’ number of classes per feature, the number of possibilities is n x m. Now a quant can build a classification model with the features as variables and the S&P 500 index as the response, and probability distributions for each possibility can be generated. Bayesian networks are a useful tool for this type of analysis. An example result of this type of model, a probability distribution table, is shown in table 1.

Figure 2: Example possibility matrix using economic topography data

Source: Author

Source: Author

Table 1: Example probability distribution table

Source: Author

Source: Author

Applications

Armed with a probability distribution table, ideally with some skewed distributions, as shown by the far right and far left columns in table 1, an analyst can then write an algorithm to implement table 1 into a trading program. Also, the node set created by using the economic topography can be written into a larger, more complex, model incorporating many other variables such investor sentiment, economic fundamentals, and technicals. Finally, the model and results are also easily implemented into analysts’ larger toolkit, which includes fundamental, and bottom up analysis.

DISCLAIMER

Any material provided in this blog is for general information use only. You should not act based solely upon the materials provided herein. Vital Data Science Inc. advises you to obtain professional advice before making investment decisions. Your use of these materials is entirely at your own risk. In no event shall Vital Data Science Inc, its officers, directors or employees be liable for any loss, costs or damages whatsoever.

Comment

Comment

Developing trading algos with Bayes Networks

 

Bayes Theorem: Google uses it to steer driverless cars, it helped solve Churchill’s greatest problem during WWII – German U-boats devastating Ally navies. Let’s see how we can use it to increase our chances of successful trades.

The idea behind Bayes Theorem is simple: use an event that has already occurred (B) to gain further insight into the future event you are interested in (A). Mathematically,

What we are interested in is the probability that going long or short a financial instrument as a result of receiving a buy or sell signal from our model will be followed by price appreciation or depreciation, respectively. In our case, the event that has already occurred is the buy or sell signal derived from fundamental or technical data. In other words, if our model is giving a buy or sell signal, what is probability that the trade will be successful?

Bayesian Networks are used to implement Bayes Theorem on data sets. A Bayesian Network is represented graphically by a series of nodes connected by arcs. Arcs represent the relationship between parent and child nodes, where a child node is dependent on the status of its parent nodes. A simple Bayesian Network is depicted below.

For the case we are interested in, parent nodes are statistics derived from fundamental and/or technical data, and the child node is price movement of the financial instrument we are modeling. Once the problem is graphically represented, an algorithm is implemented which ultimately determines the probability distributions of the child node. It is possible to use either discrete or continuous data for parents or children. Further, the network can consist of much more complex child parent relationship than depicted above and the analyst need not specify complicated datasets graphically; algorithms such as the growth-shrink algorithm can be implemented to learn the network from the data.

Once the network and probability distributions are determined, the combination of statistics that yields the highest chances of success is programmed into a trading algorithm. After careful back testing and stress testing, we are ready to implement.

Confused? Here’s an example.

Example. The Vital Data Science team used the above approach to implement a trading algorithm for GE equity. Our network, depicted below, consists of two technical/fundamental statistics as parent nodes and GE equity price movement as the child node. Each node had three discrete levels: buy, hold, sell; and

where 1’s represent buy signals, -1’s represent sell signals, and 0’s represent hold signals. The problem we are solving is to determine the probability distribution when both parents are indicating buy or sell. Our hypothesis is that the buy combination of signals in the parents will yield the highest probability of a successful long in the child node, and the sell combination in the parents will yield the highest probability of a successful short in the child node.

The chart below shows the probability distribution for each distinct combination of levels in our model. The plots we are interest in are circled green for buy and red for sell.

Image3.png

The plot tells us we’ve created a strong model. When both statistics give buy signals there is an 80% chance of positive future price movement, and when both statistics give sell signals there is a 70% chance of negative future price movement. The prevalence of positive price movement is 54% and negative price movement is 46%, therefore our model is predicting significantly better than guessing.

We back tested the approach and the chart below shows periods where we would have jumped into and out of GE equity. Note that the model is sending buy signals now, however as we’ve not received a sell signal to date we cannot assess the result of the latest round of buying/selling.

Results. How did we do? Fantastic! Between January 1, 2010 and February 2012 our strategy returned 90% vs. a buy and hold return of 20%, excluding dividends. The obvious fallacy of this strategy is that we did not participate in the significant run-up in share price between February 2012 to present. However, even without participating in the latest run-up, our model returned 12% above a buy and hold strategy.

The reason the model skipped the majority of the period between 2012 to present is because we set one of the statistics based on annual values, therefore the model was slow to adjust to the run-up in price. The simple solution is to shorten the duration of that statistic, for example to reset the statistic quarterly. That said, one has to balance the frequency of trades with the ability to predict them correctly; in this case we decided to cut our trading frequency in order to achieve higher chances of success.

DISCLAIMER

Any material provided in this blog is for general information use only. You should not act based solely upon the materials provided herein. Vital Data Science Inc. advises you to obtain professional advice before making investment decisions. Your use of these materials is entirely at your own risk. In no event shall Vital Data Science Inc, its officers, directors or employees be liable for any loss, costs or damages whatsoever.

Comment