Scenario Analysis: "What-if" Stress Testing For Advisors

Tuesday, May 10, 2011 10:14
edit
Scenario Analysis:

With the housing crash and financial crisis not fully in the rear view mirror, and new threats like an oil price spike facing us, it's no wonder that much of the investing public is scared. That fear has left investors missing the lion's share of the market runup since 2009. How can advisors overcome clients' fears and plan for future economic pitfalls?

This Website Is For Financial Professionals Only


At the heart of investors' paralysis is a fear of the unknown, a fear of the many potential doom-and-gloom scenarios facing the US economy. Advisors can potentially ease clients' fears by quantifying them, and showing clients a realistic picture of the risks facing their investments. Many advisors have access to traditional risk analysis tools like Monte Carlo simulations and VaR (value-at-risk) tools. Unfortunately, these tools failed during the financial crisis. Monte Carlo and VaR suffer from the fact that they are backward-looking models which operate within the confines of "normal" markets - so that they cannot account for black swans like those we've recently experienced. While VaR models typically calculate the impact of a 2-standard deviation move in the markets, the financial crisis was at minimum a 6-sigma (standard deviation) event, and perhaps as much as a 16-sigma event!

Scenario Analysis provides a better way to address the big picture risks facing the economy today, the kinds of risks that keep clients (and advisors) up at night. Hedge funds and Wall Street trading desks have been modeling forward-looking scenarios for some time, and use these models to assess the risks (and rewards) of major macro events. HiddenLevers has now made this technology available to investment advisors, enabling advisors to stress test clients' portfolios through a range of "what-if" scenarios.

How does HiddenLevers Scenario Analysis work? We start by defining scenarios around macro themes like stagflation, rising oil prices, Euro Zone defaults, and more. Our economic research team creates assumptions around how economic indicators like GDP, inflation, interest rates, commodity prices, and more might react in these scenarios. HiddenLevers' analysis engine determines the correlations between stocks, ETFs, and mutual funds to the economic indicators. The HiddenLevers engine can then stress-test real client portfolios in each scenario, showing advisors where potential risks lie. While no projection is perfect, this approach enables advisors to tackle clients' fears head-on by looking
at real-world scenarios.

To learn more about HiddenLevers and Scenario Analysis, visit HiddenLevers and watch our upcoming webinar here at Advisors4Advisors on May 20th.

Comments (17)

...
brentb843
Absolutely silly. This program assumes that you need to know what is the cause of the crisis, not the recognition of a 6 or 16 sigma event.

To state it a better way, did Hidden Levers have the manner to illustrate the CDS implosion prior to 2007? If it was obvious, it would not be a crisis....
brentb843 , June 30, 2011
...
hiddenlevers
Brent, HiddenLevers doesn't try to predict the future. We do provide a modeling framework based on correlation analysis, so that you can model any scenario you like, including both those that we've constructed and any custom scenario that you can devise.

This is a more flexible approach than that available in VaR or Monte Carlo, and it also exposes security-specific risks that fall outside the realm of those models.
hiddenlevers , April 03, 2012
...
brentb843
Doesn't correlation always go to 1 in most crisis?

The premise of HiddenLevers for advisors who work with clients on achieving goals is ridiculous at best. It does not matter what causes the event, it matters that it
happened.

If I know that small growth can fall 75%, then I avoid it ALWAYS, regardless of economic conditions, which rarely can be predicted.

For example, I cannot imagine an advisor saying 'sorry you're broke, I thought natural gas would rally, but instead gold collapsed.'

This methodology assumes the importance is the specific cause and that advisors must maintain a specific allocation in a portfolio. Is this not really more investment manager focused?

Silly, silly, silly.....
brentb843 , April 03, 2012
...
hiddenlevers
Brent,

As I've said before, HiddenLevers doesn't attempt to predict the future. We do attempt to model different potential downside (and upside) scenarios, but the model is open to any assumption. For instance, if you simply want to stress test with a generic downside shock of -20% on the S&P, that's easily done. But we also take into account the other correlations that can occur in that sort of scenario - that's something other approaches don't do.

We work with both active portfolio managers and advisors focusing exclusively on asset allocation, and offer different benefits to each group.

Many advisors use HiddenLevers to help their clients understand potential risks, and they tell us that clients understand this sort of risk modeling much better than generic statements on volatility and probabilities. If an advisor needs to answer clients' fears on Euro Zone debt issues, or on rising inflation, HiddenLevers provides a concrete way to do so.

That doesn't sound very silly to me.
hiddenlevers , April 03, 2012
...
hiddenlevers
Brent, historically correlations don't always go to 1 in crises - we have data on quite a few historical scenarios that shows that this assumption is false.
hiddenlevers , April 03, 2012
...
brentb843
I understand...what I am saying (again) the reason the portfolio drops does not matter. If it can be modeled, then it can be avoided, right?

Hiddenlevers is an exercise in mental masturbation....it assumes the prediction will offset the event.

The ONLY way to avoid a 20% drop in something is not to have exposure to it....you cannot reduce risk via risk modeling.

Proper risk modeling no longer relies on correlations. See Wilmott....
brentb843 , April 03, 2012
...
brentb843
Praveen, give an example. The NFP was incorrect, want to try again?

If Hiddenlevers is designed to model risk, then ok. But with the post, you are attempting to say 'buy this, model an event and avoid it.'

brentb843 , April 03, 2012
...
hiddenlevers
Brent,

In the original post I clearly stated that HiddenLevers is designed to model (assess was the word in the post) risk. We also help advisors present and explain potential risks to their clients.

If you're interested in understanding what HiddenLevers actually does, I'd be happy to provide a demonstration. Your feedback on our product, our approach, and any improvements that we can make would be welcome as well.

Here's a link to sign up for a demo if you'd like to see the product and provide feedback on it:

http://bit.ly/HFosQP
hiddenlevers , April 03, 2012
...
brentb843
Look, I'm not interest. My agenda is to show this is a waste of money. Your posts do not match your latest arguments.

The entire notion is broken....

If it is as you claim, how did you end up with a post on NFP?
brentb843 , April 03, 2012
...
hiddenlevers
Brent, I realize that you're not interested in HiddenLevers - but I thought perhaps you'd like to see the product so that you could understand what you're commenting upon. If you have a few minutes to call me at 404-941-8600, I'd like to understand your perspective.

The post on NFP is also clear, and factually accurate. In our articles here we like to point out correlations that exist, and data that we think is important.
hiddenlevers , April 03, 2012
...
agluck
Civility, Brent! Please. Hidden Levers has satisfied customers -- advisors pleased with its solution, undermining your harsh judgment.
agluck , April 03, 2012
...
brentb843
Andy - I understand that they are a sponsor, but I got frustrated with Praveen saying one thing and posting another.

I do not need to see HiddenLevers as I reject the notion that prediction of economic scenarios matter.

Praveen continues to say, when called out, that 'no, HiddenLevers is not predicting, just showing risk,' then he writes an absurd and inaccurate post about NFP being 'the indicator.'

Is there evidence that the advisors using HiddenLevers providing better performance towards achieving investor goals?

Its not the product, it is the notion that predicting economic scenarios matter. Advisors claim through marketing (heck Andy, you set up websites) that they are meeting goals of investors.

Come on, I am fine with Praveen marketing that HiddenLevers can model scenarios, but he is selling 'indicators.'

What is more surprising, and I would love to perform research is how an advisor working towards achieving an investor's goal is remotely utilizing this.
brentb843 , April 03, 2012
...
brentb843
Its about goals, not markets.....

As you know Andy, I have no agenda as I am building an application to let investors see if their advisor can build portfolios towards achieving goals. Not selling something to advisors.

To finds value in this methodology, the following assumptions must be made:

That retirement income can be achieved through relative outperformance of a market index.

That advisors, who work with investors, have the time and energy to become macro economists.

That the reasaon the account falls 20% is more critical than the exposure to the 20% in the first place.

Crisis can be modeled and then avoided.

Modelling risk is not the same as avoiding it.
brentb843 , April 03, 2012
...
hiddenlevers
Brent,

I apologize if I made it seem like HiddenLevers was "selling indicators." Actually we provide the articles on correlations and indicators as free informational content, as I can't exactly run a scenario model against a client's portfolio on a blog post, can I? We write on scenarios and economic topics that we think are of interest to advisors.

I know you have no interest in the product, but here are some links so that you can see what it is for yourself:

https://www.hiddenlevers.com/about-hiddenlevers/

https://www.hiddenlevers.com/hiddenlevers-how-it-works/
hiddenlevers , April 03, 2012
...
brentb843
Lets look at this from the perspective of advisor marketing. Many advisors claim a financial plan is much like a blueprint when building a home. So, I live on the coast in South Carolina which is both hurricane and earthquake prone.

I sit down with my architect and he has 40 different plans, each one attempting to model if/when/what cause would have an earthquake or hurricane occur. He shows me the data that there has never been an earthquake or hurricane at the same time, so I need to choose whether to build to protect against an earthquake or hurricane.

Not being stupid, I fire the architect and hire one that will build my home to code. Then fire burns it down....

The only way to avoid a hurricane or earthquake is to move, not to model. And some risks like fire are always prevalent.
brentb843 , April 03, 2012
...
agluck
For tthe record, Brent, Hidden Levers is not a sponsor. Sponsorerd postys on A4A are tagged as "Sponsored Content." Please take more care to avoid making incorrect assumptions and mistating facts.
agluck , April 04, 2012
...
brentb843
Sorry Andy - you are correct to call me out.
brentb843 , April 05, 2012

Write comment

You must be logged in to post a comment. Please register if you do not have an account yet.

busy