Any sort of bias present in the training data will be replicated in the model. If the police are biased in whom they target, that group will naturally show a higher crime rate. Which would easily be picked up in any sort of statistical model. Leading to a biased model.
The drop in crime happened nationwide during this same time in places which did not include NYC. [1] The timing of this seemed to have been more coincidental with the drop in crime.
I have worked as a health actuary for the past 6 six years with degrees in both math and computer science. I am happy with my career choice. Prior to this I did work several years as a programmer and decided while I enjoy programming it is not something I want to do all day every day.
My current position allows me to make use of some of those skills along with learning a wide variety of other skills. Although from the sounds of other people's posts I might be lucky in my current position, where I do have variety and challenges and more than only routine work.
But from an education stand point I am glad that I went the route that I did given the additional skills rather than an actuarial science degree. The one downside is that it has made the exam process a little longer.
The cost of insurance is largely driven by the underlying cost of what you are insuring. Unless they are able to lower the cost of care significantly, the price difference between a non-profit and for-profit is marginal.
The measure by itself is not sufficient. Which is why all the additional analysis was needed.
1. Using the current district map the last set of elections show that Wisconsin had a large gap.
2. Compared to other state's the gap is an outlier.
3. By creating a large number of alternate maps within the state satisfying all the other requirements that gap was still an outlier.
4. Calculating the gap under different voting outcomes showed the result to be robust even under a 5 point swing to the democrates. (This is where the discontinuity would show up if there results were not robust.)
I am familiar with NY medicaid. They do publish a way to calculate the Medicaid default rate. Insurers do not have to pay exactly this but it provides a decent base line. Here is a basic description of how inpatient pricing works.
Each year the state publishes the set of hospital rates and intensity weights for each DRG (Diagnosis-Related Group) and severity combo (currently using weights developed in 2014). So a DRG of 460 (Renal Failure) with a severity 2 has a weight of 0.7393. Now the actual cost will depend on which hospital you go to since each hospital has a different base rate. For example each Mount Sinai hospital has a base rate of $8,743.45 while Niagara Falls memorial hospital has a base rate of $5,558.99. Each hospital also has a per discharge rate. To calculate the default rate take the hospital base rate x DRG intensity weight + per discharge rate.
Mostly due to all of the variables that go into pricing a claim. And that logic on lives in the insurer's claim processing system.
Pieces that can impact the price. Your insurer and what product you have. These will affect who is considered in-network and the fee schedule to use. Different insures will have different arrangements. Depending on the product if you have a narrow network product they may or may not be in-network. It could also depend on the location. A provider can be in-network in one location but not in another.
Also the procedure that is actually performed may be slightly different from what was planned due to unforeseen circumstances.
This is assuming the provider is aware of what the actual costs are. In many cases they don't even know the ballpark price since that is not the portion that they deal with.