From boom to bust: Searching for sustainability in the new oil climate

Times have changed. As the oil industry shutters and sheds investments that made sense during the two-year period in which oil rode comfortably above $90, the market is establishing a new equilibrium at $40/barrel. This despite the fact that the Baker Hughes domestic rig count is down 64 percent. It’s an indication of how big the US shale oil boom was (the US Energy Information Administration reckons it was the largest expansion in American crude oil production in more than a century) – and how tenacious the year-long bust has turned out to be.

One clear indication of the current oversupply is how the price of crude reacts to news headlines. There is no doubt that a major terrorist attack as horrific as the tragic events in Paris will have serious consequences for the Middle East. In the past, any incident that increased tensions in oil-producing regions resulted in an immediate, substantial market reaction. But in the present market, oil prices did not respond with a run-up – a sign not of indifference, but of glut. With inventories full to bursting, it's news of drawdowns – rather than events that could spark supply shortages – that provides short-term stimulation in the market.

The market impact of lower volatility and tighter OPEC controls is reflected in the flatness seen in the latest VirtualOil simulation (see Fig. 1 below). Value is riding small sparks of occasional market volatility, and Value-at-Risk is closely tracking that dollar-per-barrel Mark to Market valuation. The recent decision to restructure VirtualOil at a $25 strike price means the portfolio more closely reflects many producers’ current level of operating costs. The difference is that our fictitious oil portfolio can be nimble because it's structured around derivatives, so it can generate profit in the current price environment. VirtualOil’s options are well into the money, delivering substantial positive cashflow. But many nonfictional producers’ results are saddled with additional overhead and capital expenditures, meaning they're losing money on each barrel produced.

VirtualOil Rolling Five-Year Portfolio

Fig. 1: VirtualOil Rolling Five-Year Portfolio, Nov 2015

Producers are focused on hanging in there for another year or more of lower prices. That means optimizing current production using analytics. SAS has been working with customers to use analytics to draw optimization out of the data streaming off the oilfield to create more efficient use of assets in the low-price environment.

One example is analysis of downhole sensor data in wells employing steam-assisted gravity drainage (SAGD) to produce heavy oil and bitumen. Using analytics to achieve the optimal play between injected steam and production flow can have a significant effect on the cost of producing a barrel of oil.

Spot and near-term prices have come off and there’s a steep contango on the forward curve. WTI is up $7 one year out, suggesting the market sees some upside ahead. The key word for producers is sustainability.

The hypothetical derivatives-based oil production firm VirtualOil simulates the performance of a generic crude oil asset, and delivers sectorial exposure to the commodity oil market. The reorganized VirtualOil structure starts up with an investment of $200MM in monthly average price call options with a strike price of $25 per barrel on the price of West Texas Intermediate (WTI) light sweet crude oil. The strip of options starts at 10,000 barrels per day and extends out for five years with a 20 percent average annual decline in underlying notional barrels, replicating a physical oil asset. VirtualOil initially holds notional crude oil reserves of approx. 10MM barrels. Monthly cash flow is generated when the daily average WTI price relative to the preceding month exceeds $25 per barrel. Cash flow is reinvested monthly at 5% and the project winds up when the reserves are depleted.

See additional simulations below. VirtualOil is managed in SAS® BookRunner with reports surfaced in SAS® Visual Analytics. Learn more about SAS® BookRunner’s state-of-the-art commodity trading and risk management capabilities. 

Disclaimer: This is a fictitious portfolio and is not a solicitation to trade.

VirtualOil Jan 15 portfolio

Fig. 2 VirtualOil Jan 15 portfolio






VirtualOil Jan 2014 portfolio

Fig. 3: VirtualOil Jan 2014 portfolio

Post a Comment

Data science doesn’t have to be difficult

You could argue that it’s misguided for someone like me to say data science doesn't have to be difficult. After all, I’ve been in the industry for many years and should have a few tricks up my sleeve for dealing with data. But with the latest data visualisation technology – it even includes built-in analytics capabilities - you don’t have to be an expert to use it.

I saw this first-hand when one of our own interns used SAS Visual Analytics during his time with us earlier this year. In just a month, he delivered a full analysis of how UK universities are using data analytics for research and academic purposes.Tightrope

Each summer, we recruit eight to 10 students for a SAS summer internship that gives them a taste of being a data scientist. Over a six-week period, our interns attend a short training course and are assigned to different business departments. Each intern is tasked with solving a unique business challenge using SAS data analytics techniques.

Here’s what Kushal Shah shared with us during his internship: Read More »

Post a Comment

What skills do you need to be a data scientist?

To be successful as a data scientist you need technical skills like programming and mathematical skills, but you also need passion and the ability to put information into context and explain its significance, says Dr. Goutam Chakraborty of Oklahoma State University.

In the video below, Chakraborty explains that Oklahoma State University has launched a new MS in analytics program to train data scientists who will graduate with big data and data science certifications.

Hear more from Chakraborty, including how to apply your data science skills:

And if you're looking to hire a data scientist with the skills Chakraborty describes, be sure to read our e-book, Your data scientist hiring guide, which includes 20 interview questions and three data scientist profiles.

Post a Comment

How to future proof your data strategy

In my previous post, I discussed some of the challenges and costs organizations face when they’re stuck in Excel hell with no real data strategy. Now that we’ve discussed the problem, let’s dive into the solution.

Every organization needs a data strategy with these building blocks:strategy

Your top priority is linking your data to company strategy. Here are the steps you need to take:

  1. Identify strategic goals - i.e. long-term company objectives.
  2. Break the strategic goal down into objectives. Assign a stakeholder to each objective as the driver/owner of that objective and define initiatives that fall under each objective.
  3. Create a project work group that includes IT, end-users and information consumers. This small work group will start the project(s) that contribute to the larger initiative. With your first few projects, go for quick wins (projects that show maximum value for a minimum of effort). Early successes will help prove the value of the overall initiative and help get (and keep) the team motivated.
  4. Translate projects into solutions. The cross-functional project team will work together to address challenges from multiple angles and translate business requirements into actual solutions.

Read More »

Post a Comment

Let consumers teach you a thing or two this holiday season

The holiday season is a festive time of the year. But it can also be a nail biter for retailers. Months of planning can be sidetracked by fickle consumers or aggressive pricing from competitors.

Paying attention to even a little bit of data can work wonders.

In October, SAS conducted online research among adult consumers from Australia, Canada, New Zealand, the United Kingdom and the United States to uncover their holiday buying plans. This survey won’t help you restock your stores or redo your promotional plans, but it points out the potential of what you can do with your own data now and for holiday seasons to come.

We talked to 3,458 consumers. As a retailer, you’ve got lots more consumer information than that at your fingertips already. So what are you doing with it?

Here are four insights from the survey and some thoughts on how you can enhance these insights for your own company. Read More »

Post a Comment

Best practices for using data and analytics

Know your data. Do a needs analysis. Organize for success. Empower users. These are four best practices for data and analytics that you'll want to hear more about.

In my first three posts in the Analytics in Real Life blog series, we learned how higher education customers are using SAS and why they chose SAS. These customers also explained the positive impact of using SAS and analytics for their users and institution and they also shared tips for gaining buy-in for data and analytics projects.

In this last post in the analytics in real life blog series, the following customers share best practices for using data and analytics. Read More »

Post a Comment

Sawdust and SAS Studio: Thoughts on a liberal arts education during an IoT workshop

I brushed aside some sawdust on the workbench and set my laptop down. It wasn’t really mine. SAS Library Services had kindly lent me a new laptop for the “Making Sense of Sensor Data” workshop at UNC’s BEaM Makerspace. I had just set the laptop down…in sawdust. Like any normal person, I immediately began to think of a good lie I could tell when I returned the laptop.

Two students learning about arduino, sensor data and SAS

Two students learning about Arduino, sensor data and SAS.

SAS: Why does this laptop smell like sawdust?

ME: I don’t know. I noticed that, too. I think it must be Windows 10.

SAS: Windows 10?

ME: Yes, I think Windows 10 just smells like that.

SAS: But we don’t have Windows 10 on that laptop.

ME: No, of course not. It kept asking me if I wanted to install Windows 10 and I kept clicking “no” and then I smelled something like burnt leaves…

SAS: Windows 10?

ME: Yup. I think I read about this online somewhere. They call it the “Winter Woodburn” scented desktop.

That’s the problem with lying. There are a small group of people who are exceptionally good at it and the rest of us are lousy. I decided I would defend myself with the truth.

“Yes, I set the laptop down in some sawdust,” I would say, “but at least I didn’t set it down beside the table saw.” That would earn me some points back, I thought.  Resolved, I pressed the power button to turn the laptop on and proceeded as planned. To my relief, the loaner laptop was apparently impervious to a little sawdust. Read More »

Post a Comment

The hell of Excel, or why you need to future proof your data strategy

This summer, I had several interesting sessions with customers and prospects. Much to my surprise, two of them, both multinational organizations, were doing most of their data related tasks in Excel.

This happens every now and then -- I come across organizations (like yours?) where people are manipulating and ‘analysing’ data in Excel. In many cases, they’re doing this even though other, much better, tools are available.

When I see what some of them have accomplished, I must admit it’s astonishing. They build extremely complex Excel sheets full of formulas, v-lookups and macros that do their tricks.

But when I start asking questions, it quickly becomes clear that even though Excel does the things they want it to do, the time and effort it takes far outweighs the business value they get in return.

This is what I like to call ‘the hell of Excel’ -- a lot of people are experiencing it (and some don’t even know they’re in it). Read More »

Post a Comment

Text mining, the movie

42-15326326I hope by now you’ve seen the movie “The Imitation Game” with Benedict Cumberbatch and Keira Knightley. It’s the true story of Alan Turing, whom many consider to be the father of the modern-day computer and the discipline of computer science.

Turing’s innovation (and the movie) takes place in early 1940’s London during the Nazi domination of Western Europe and the Atlantic. The Germans were killing three allies every 15 seconds in the early years of the war.

Like all military organizations, the Germans were communicating military instructions via coded messages. These messages were sent and received by a code machine nicknamed “The Enigma.” The Enigma codes changed every 24 hours. While the Allies were able to intercept all German messages, they couldn’t decrypt them, thus rendering the messages useless.

The inability to decipher German messages was preventing the Allies from gaining any military intelligence -- and that was costing them the war.

Read More »

Post a Comment

3D printing: yes or no?

3D printer printing a test print

A 3D printer test print.

By now, you’ve probably seen a video of a 3D printer discharging layers of plastic to create a model of a building or a plastic figurine. You may have heard stories about 3D printed guns, 3D printed airplane parts and even 3D printed body parts parts.

While 3D printers are becoming more common, they are still a long way from being common household or break room devices.

How will 3D printers be relevant in the field of analytics? And what are their true strengths? We asked Matthew Horn, manager of SAS’ Emerging technologies UI lab for his thoughts – and here’s what we found out.

What it is

3D printing is a method for creating physical objects or models by “printing” multiple thin layers of materiel successively to form an object. 3D printers can easily print intricate shapes and interlocking pieces. Read More »

Post a Comment