We're looking for a Deputy Commissioner of Operations

We're currently recruiting for a Deputy Commissioner of Operations. This in an executive level exempt position that manages 35 employees in the following areas: Human resources, budget and fiscal, facilities and telecommunications, information technology and public records.

The successful candidate will be an active member of the Executive Management Team (EMT), setting the strategic direction of the agency, developing legislative priorities, ensuring fiscal responsibility and creating an inclusive, performance-based work environment.

Here's the full job announcement. Please share with anyone you think could be interested.

The job is open until Dec. 9, 2014.

Not only verbs but also believes can be conjugated

Following on from last week, where I presented a simple example of a Bayesian network with discrete probabilities to predict the number of claims for a motor insurance customer, I will look at continuos probability distributions today. Here I follow example 16.17 in Loss Models: From Data to Decisions [1].

Suppose there is a class of risks that incurs random losses following an exponential distribution (density \(f(x) = \Theta {e}^{- \Theta x}\)) with mean \(1/\Theta\). Further, I believe that \(\Theta\) varies according to a gamma distribution (density \(f(x)= \frac{\beta^\alpha}{\Gamma(\alpha)} x^{\alpha \,-\, 1} e^{- \beta x } \)) with shape \(\alpha=4\) and rate \(\beta=1000\).

In the same way as I had good and bad driver in my previous post, here I have clients with different characteristics, reflected by the gamma distribution. I shall call the gamma distribution with the above parameters my prior parameter distribution and the exponential distribution the prior predictive distribution.

The textbook tells me that the unconditional mixed distribution of an exponential distribution with parameter \(\Theta\), whereby \(\Theta\) has a gamma distribution, is a Pareto II distribution (density \(f(x) = \frac{\alpha \beta^\alpha}{(x+\beta)^{\alpha+1}}\)) with parameters \(\alpha,\, \beta\). Its k-th moment is given in the general case by
E[X^k] = \frac{\beta^k\Gamma(k+1)\Gamma(\alpha - k)}{\Gamma(\alpha)},\; -1 < k < \alpha. \] Thus, I can calculate the prior expected loss (\(k=1\)) as \(\frac{\beta}{\alpha-1}=\,\)333.33.
Now suppose I have three independent observations, namely losses of $100, $950 and $450 over the last 3 years. The mean loss is $500, which is higher than the $333.33 of my model.

Question: How should I update my belief about the client's risk profile to predict the expected loss cost for year 4 given those 3 observations?

Visually I can regard this scenario as a graph, with evidence set for years 1 to 3 that I want to propagate through to year 4.

Read more »

The Defense Rests

It is exhausting.  This choice.  This job.  This desire to craft a non-ideological, pragmatic path is taking all of my energy.  Confronted daily by people who either cannot, or will not, see the full picture that is the Patient Protection and Affordable Care Act (PPACA or Obamacare), I find myself calming the fears of its detractors or clarifying the rules to its biggest supporters.  To the right, and the righter than right, I find myself defending the law, or at the very least, the need for change.  Defending the insurers from the left could be a full time job in of itself.

Randy (name changed) called Friday.  He wanted to know when they were going to cancel his group health insurance policy.

Why would Medical Mutual cancel your company’s insurance?

Because my policy doesn’t cover any women or children.

But you don’t have any female employees, right?

Yeah, but they’re gonna cancel me!

OY, Randy, you’ve been watching FOX again.

It took fifteen minutes to reassure Randy.  Now, no one on FOX really said that a small business would lose its health insurance if there weren't any women or children on the policy.  That’s silly.  But the daily barrage of negativity, conspiracy theories, and half-truths take their toll on the viewers.  One day you’re a concerned business owner.  The next you are trying to get one of your employees to get married so that you can retain your group coverage.

It doesn’t get any better on MSNBC.  With neither an ounce of irony nor embarrassment, the outpost of the left gives us Howard Dean, the former governor of Vermont.  I’m sure that an extensive GOOGLE search might find an instance when Gov. Dean knew what he was talking about.  I’m just positive that none of his pronouncements about health insurance or the PPACA have any basis in reality.

For example, Governor Dean was recently discussing the disastrous roll-out and the policy cancellations.  He was on Morning Joe and several other shows.  He opined that all the President had to do was to hire a bunch of unemployed kids, put them in a call center, and have them ring up everyone whose policy had been cancelled.  The kids could enroll everyone into Obamacare!

I doubt that approach would be welcome in Vermont, a state with less than half the population of Greater Cleveland.  I know that wouldn’t fly here.  Who explains the policies to the “kids”?  Vermont may have only one or two insurers and only a few options, but Ohio, California, and any other state that has an actual city or two will have multiple insurers and dozens of choices.  But on one has ever explained insurance, or economics, or city life to Governor Dean.  And there is absolutely no reason to do so now.

40 vs. 5

The PPACA was sold to the American public as a universal win.  Everyone would get better, more comprehensive health insurance for a lower monthly premium.  This blog has repeatedly pointed out that that was not possible.  The airwaves are now filled with the horror stories of cancelled policies and jacked-up premiums.  The right emphasizes every problem, real or imagined.  The left has a new argument – Isn’t it OK to inconvenience five million people so that 40 MILLION AMERICANS can now get access to affordable health care?

What a bunch of hooey.  This wasn’t a Hobson’s choice.  It wasn’t remake our entire system or do nothing to help the uninsured and the under-insured in our country.  We could have accomplished much of that goal by expanding Medicaid.  You don’t score many points by minimizing someone else’s loss.

I have watched my words parsed in the comment sections of Facebook and the AOL Patch.  The attorneys and attorney wanna-be’s who troll for fights can’t tolerate civil discussions.  One guy was convinced that all insurers will cancel their clients at the first sign of a claim.  Another reader is positive that the PPACA is the harbinger of the Apocalypse.  The extremes are so extreme.  The middle is lonely and damn near empty.

For the record:
  • Insurers pay claims.  My clients have benefited from their coverage.
  • The status quo was not sustainable.
  • There is a kernel of truth in everything you see and hear on FOX and MSNBC.  You need more than kernels.  You need a meal.
Take a deep breath.  We will all get through this together.  But for the moment, the defense rests.



Predicting claims with a Bayesian network

Here is a little Bayesian Network to predict the claims for two different types of drivers over the next year, see also example 16.15 in [1].

Let's assume there are good and bad drivers. The probabilities that a good driver will have 0, 1 or 2 claims in any given year are set to 70%, 20% and 10%, while for bad drivers the probabilities are 50%, 30% and 20% respectively.

Further I assume that 75% of all drivers are good drivers and only 25% would be classified as bad drivers. Therefore the average number of claims per policyholder across the whole customer base would be:
0.75*(0*0.7 + 1*0.2 + 2*0.1) + 0.25*(0*0.5 + 1*0.3 + 2*0.2) = 0.475
Now a customer of two years asks for his renewal. Suppose he had no claims in the first year and one claim last year, how many claims should I predict for next year? Or in other words, how much credibility should I give him?

To answer the above question I present the data here as a Bayesian Network using the gRain package [2]. I start with the contingency probability tables for the driver type and the conditional probabilities for 0, 1 and 2 claims in year 1 and 2. As I assume independence between the years I set the same probabilities. I can now review my model as a mosaic plot (above) and as a graph (below) as well.

Next, I set the client's evidence (0 claims in year one and 1 claim in year two) and propagate these back through my network to estimate the probabilities that the customer is either a good (73.68%) or a bad (26.32%) driver. Knowing that a good driver has on overage 0.4 claims a year and a bad driver 0.7 claims I predict the number of claims for my customer with the given claims history as 0.4789.

Alternatively I could have added a third node for year 3 and queried the network for the probabilities of 0, 1 or 2 claims given that the customer had zero claims in year 1 and one claim in year 2. The sum product of the number of claims and probabilities gives me again an expected claims number of 0.4789.


[1] Klugman, S. A., Panjer, H. H. & Willmot, G. E. (2004), Loss Models: From Data to Decisions, Wiley Series in Proability and Statistics.

[2] Søren Højsgaard (2012). Graphical Independence Networks with the gRain Package for R. Journal of Statistical Software, 46(10), 1-26. URL http://www.jstatsoft.org/v46/i10/

Session Info

R version 3.0.2 (2013-09-25)
Platform: x86_64-apple-darwin10.8.0 (64-bit)

[1] en_GB.UTF-8/en_GB.UTF-8/en_GB.UTF-8/C/en_GB.UTF-8/en_GB.UTF-8

attached base packages:
[1] grid stats graphics grDevices utils datasets methods
[8] base

other attached packages:
[1] Rgraphviz_2.6.0 gRain_1.2-2 gRbase_1.6-12 graph_1.40.0

loaded via a namespace (and not attached):
[1] BiocGenerics_0.8.0 igraph_0.6.6 lattice_0.20-24 Matrix_1.1-0
[5] parallel_3.0.2 RBGL_1.38.0 stats4_3.0.2 tools_3.0.2

Why the commissioner decided against allowing canceled policies back into Washington

President Obama’s announcement this week that previously canceled individual health-insurance policies for 2014 could be reinstated – at the discretion of insurance commissioners – drew a lot of attention.

Washington state Insurance Commissioner Mike Kreidler acted quickly, deciding against allowing the previously discontinued policies to be put back into effect. The reasons are straightforward: The proposal does not make sense for Washington because of the overall negative effect on the stability of the health-insurance market.

To learn more about Commissioner Kreidler’s decision, please check his statement.

You can also see what is really happening with those canceled policies and how many consumers can qualify for more comprehensive coverage at less cost.

Where Do They Bury The Survivors?

A plane crashes on the Mexican – United States border. On board were U.S. citizens, Mexicans, Brazilians, and three passengers from Argentina. Where so they bury the survivors?

It is a classic misdirect. A mental sleight of hand. You know the answer. We don’t bury survivors.

Sleight of hand is an art. The best practitioners can shake your hand while they lift your watch and wallet. The trick for us is to watch them at work without becoming a victim.

I recently received an urgent email. A client forwarded Newt Gingrich’s article, “Obamacare’s Marriage Penalty and Divorce Incentive.” Was this true? Is the President anti-marriage?

Newt Gingrich as the defender of the sanctity of marriage? Guard your wallet!

I won’t bore you with the numbers. The federal subsidies are based on the size of the family, the ages of the insured, and are factored on the federal poverty level. A married couple with one child doesn’t need 33% more income when they have a second child. Couples don’t need twice as much income as singles to pay for food and shelter.

Will a few people get divorced to get a bigger health insurance subsidy? Perhaps. Of course, that also means that these people will pay more state and federal income tax. These things have a way of balancing out in the end.

The bottom line is that Newt Gingrich knows that this is irrelevant. Newt would be campaigning against government waste and another inefficient entitlement program if the subsidies were calculated differently. It is just a sleight of hand.

We should, by now, be used to this from our politicians. Sometimes it is a mental misdirect. Sometimes it is a misstatement. And there are other instances when the politician tells the truth, technically, but what he/she said and what we heard are not even closely the same.

Example? My favorite is “If you like your policy, you can keep it. Period”. Balderdash.

I wasn’t in the room when the President and his advisors crafted that perfect slogan. “If you like your policy, you can keep it. Period.” So clear. So emphatic. So wrong. Did the President and his advisors intentionally mislead the country, or more likely, did we have a room full of people who had no idea what they were talking about?
  • Over 80% of Americans get their insurance coverage through work. If you are one of them, you don’t choose your plan, your employer does. It is not up to you.
  • Only policies on the books prior to the passage of the law and unchanged since that day, March 23, 2010, are grandfathered.
  • How long can the insurers run two separate sets of books? Policies issued prior to March 2010 have one set of rules while new policies have another. Who pays the additional cost to maintain the old contracts and monitor compliance?
This blog has tackled the grandfather issue repeatedly since August 2, 2010’s, Don’t Cry Uncle, Stay Grandfathered. Retroactive rules. Contradictory edicts. Those of us who actually work in the insurance business knew that we would see very few individual or small group policies limp across the finish line on January 1, 2014.

The President is shocked that many Americans are now losing their current policies and being forced into new, more expensive contracts.

There are some awful policies on the market that will disappear on January 1st. There are some policies that have a $25,000 or $50,000 cap. Those plans were cheap, but they did not really cover a major illness. However, about 11 million Americans are covered by comprehensive policies that will be cancelled in the next twelve months. These plans don’t conform to the new rules.

My policy is scheduled to end a year from now. Why? Because it doesn’t cover me for maternity. If nothing changes in the next twelve months, my premium will more than double next December. Of course I like my policy. And no, I can’t keep it.

The Patient Protection and Affordable Care Act (PPACA) will, eventually, help many Americans. But it would be foolish to dismiss out of hand those people who are angry or upset. You can’t just bury their fears with the survivors.


Picture from the Passen Law Group www.passenlaw.com

googleVis 0.4.7 with RStudio integration on CRAN

In my previous post, I presented a preview version of googleVis that provided an integration with RStudio's Viewer pane (introduced with version 0.98.441).

Over 80% in my little survey favoured the new default output mechanism of googleVis within RStudio. Hence, I uploaded googleVis 0.4.7 on CRAN over the weekend.

However, there were also some thoughtful comments, which suggested that the RStudio Viewer pane is not always the best option. Indeed, Flash charts and gvisMerge output will still be displayed in your default browser, but also if you work on larger charts and with smaller screen, then the browser might still be the better option compared to the Viewer pane - of course you can launch the browser from the Viewer pane as well.

Hence, googleVis gained a new option 'googleVis.viewer' that controls the default output of the googleVis plot method. On package load it is set to getOption("viewer") and if you use RStudio, then its viewer pane will be used for displaying non-Flash and un-merged charts. You can set options("googleVis.viewer" = NULL) and the googleVis plot function will open all output in the default browser again. Thanks to J.J. from RStudio for the tip.

The screen shot below shows a geo chart within the RStudio Viewer pane of the
devastating typhoon track of Haiyan that hit Southeast Asia last week.

Session Info

RStudio v0.98.456 and R version 3.0.2 (2013-09-25)
Platform: x86_64-apple-darwin10.8.0 (64-bit)

[1] en_GB.UTF-8/en_GB.UTF-8/en_GB.UTF-8/C/en_GB.UTF-8/en_GB.UTF-8

attached base packages:
[1] stats graphics grDevices utils datasets methods
[7] base

other attached packages:
[1] googleVis_0.4.7 XML_3.95-0.2

loaded via a namespace (and not attached):
[1] RJSONIO_1.0-3 tools_3.0.2

Canceled health plans? What's really happening.

There's been a lot of news lately about insurers cancelling some health plans and changing others. Consumers have been getting the word through  'discontinuation' notices.

So what's really going on? This is happening in the individual market - where people who don't get coverage from their employer - buy their own health plans.   Most likely, the existing plans being canceled or changed, failed to meet new federal standards for benefits.  The reality about most of these previous plans is that they provided extremely limited benefits - no maternity care or coverage for prescription drugs, for example.

Beginning Jan. 1, 2014, plans must provide everyone basic essential benefits, such as maternity care, prescription drug coverage and mental health services.  And no longer can insurance companies ask you about previous illnesses you may have had. They must accept everyone.

Premiums will change for some plans. But if you received a notice from your insurance company, it does not mean you have to stay with whatever they offer. You can shop around for plans, either in the new Exchange - www.wahealthplanfinder.org or from insurers selling outside of the Exchange.

Here's a map of all individual plans available by county.

For more details about insurers cancelling policies, check out this piece by Kaiser Health News.

Display googleVis charts within RStudio

The preview version 0.98.441 of RStudio introduced a new viewer pane to render local web content and with that it allows me to display googleVis charts within RStudio rather than in a separate browser window.

I think this is a rather nice feature and hence I have updated the plot method in googleVis to use the RStudio viewer pane as the default output. If you use another editor, or if the plot is using one of the Flash based charts, then the browser is still the default display.

The behaviour can also be controlled via the option viewer. Set options("viewer"=NULL) and googleVis will plot all output in the browser again.

Of course shiny apps can also run in the viewer pane. Here is the example of the renderGvis help page of googleVis. For more information about the new viewer pane see the online RStudio documentation.

For the time being you can get the next version 0.4.6 of googleVis from our project site only. Please get in touch if you find any issues or bugs with this version, or add them to our issues list.

Is this a step in the right direction? Please use the voting buttons below.

Session Info

R Under development (unstable) (2013-10-25 r64109)
Platform: x86_64-apple-darwin10.8.0 (64-bit)

[1] en_GB.UTF-8/en_GB.UTF-8/en_GB.UTF-8/C/en_GB.UTF-8/en_GB.UTF-8

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] googleVis_0.4.6

loaded via a namespace (and not attached):
[1] RJSONIO_1.0-3 tools_3.1.0

"Where do I get a detailed review of my auto and homeowners premiums?"

We recommend that you get a policy-specific premium breakdown directly from your agent (or insurer, if you buy direct). Specifically, you might want to ask for a rate worksheet comparison between your last year's premium and your current annual premium. That's a good way to get an apples-to-apples comparison of what your rates are doing.

While you're at it, it's a good idea to make sure that the type and level of coverage is what you want. If you're driving a new car, you'll likely want comprehensive and collision coverage as well as liability (the latter is required by law.) If you're driving a $1,000 car, on the other hand, maybe liability coverage is enough. These are good things to discuss with your agent or company, particularly if you have changes in your life (new car, change in household members, moving, etc.)


Frank (name changed) is a successful attorney in his early 40’s. He lives in Greater Cleveland. On Friday he completed his health insurance application, hit the submit button, and watched the website crash. This message, in bold red and blue, flashed onto his screen:
Application Wizard
  • We apologize but there was an issue with our system when submitting your application. If we were able to process your application, we will send you an email within the next hour and you will not need to do anything else. If you do not receive an email from us within the next hour, then please return to this site and submit the application to us again.
Yes, purchasing a health insurance policy in 2013 can, at times, appear daunting. But Frank wasn’t dealing with the federal exchange. The computer issue had nothing to do with the Patient Protection and Affordable Care Act (PPACA). Frank was submitting an application for a 2013 policy with Anthem Blue Cross.

Alert the media! Contact Rush! Websites crash or are shut down for routine maintenance. No one died and no one got fired.

Anthem’s producer/applicant portal was down for about six hours. Frank’s application has been accepted and by the time you read this, he may have been approved.

The exchange roll out has been a mess. Even insurance agents, professionals long familiar with the complexity of working with insurers, the government, and the public, were surprised at how unprepared the government was for this vast undertaking. We even had advance warning. Agents have had eight years of Medicare Part D (Rx) and Medicare Advantage training, seven to nine wasted hours each August. And this year we were treated to the special training classes and tests to sell on the federally run exchange as detailed in You Put Your Left Foot In. CMM (Centers for Medicare and Medicaid Services), the agency administering the agent authorization process, is still having website issues.

This is a program designed by bureaucrats who know how to make things complicated but may have no idea how to make things work.

What has been lost in this P.R. disaster is just how irrelevant all of this has been. The most important thing to remember is that the new policies don’t start until January 1, 2014. Applications accepted on October 1st or December 10th still have the same effective date. Anthem Blue Cross is not going to run out of policies. You don’t need to be the first in line.

Of course, everyone with an ax to grind has jumped into the discussion. Online insurance marketers have “volunteered” to save the day and take over the process. If the federal government would only suspend common sense and a myriad of state and federal laws, an e-marketer could corner the market and restore order. Those offers have landed with a thud.

Equally self-serving have been the Republican members of Congress who have complained about the problems their constituents are having with the roll out. You can’t spend three and a half years actively trying to sabotage a program and then complain when it doesn’t work perfectly. And please, don’t shed tears for the sick and uninsured who are having difficulty enrolling in the now available coverage. There isn’t a Republican plan to cover any of these people.

There is a bi-partisan support to bump back the “Individual Mandate” for another year. Why force people to purchase insurance if the website is difficult to access? I’m surprised that this hasn’t already happened. Everybody wins.
  • The Republicans score a moral victory. Being against anything President Obama favors enhances their campaign donations. Winning a meaningless battle shows activity.
  • The Democrats show flexibility and are allowed, once again, to play the part of the martyr.
  • Big business, the unions, and the insurers see this as one step closer to a Single Payer system. By now you are sick of hearing TV commentators talking about the Young Invincibles, the young, healthy Americans who must be forced into the system for it to work. If healthy people don’t sign up, the system devolves into a Death Spiral and implodes under its own weight.
Delaying, or worse, eliminating the Individual Mandate hastens the conversion to a Single Payer system.

The exchange website and all of its attendant issues aren’t our biggest challenge as a country. It is our lack of intellectual honesty that will be our undoing. Disorganized and unprepared, we are heading for a crash.



Our home was damaged by a windstorm. We're worried. What's next?

We get this question a lot. It's very important that you try to safeguard your home from further damage. Depending on the type of damage, you may be able to safely do this yourself or you may need to hire someone - especially if you need to get a tarp on the roof if you're dealing with utilities or damaged and unstable structures.
You'll also want to contact your insurance agent or your insurer directly to let them know what’s going on, and to get any necessary instructions from them. They can also tell you how your coverage will apply. Most importantly, stay safe!