Description: The newer predictive modeling tools are powerful but they work best when supported by a lot of computing power. Parallel processing is very powerful but often very difficult to implement. Dihui and Richard take us through the maze of options with H2O, SparkNet/MLib and Mahout. Which do you use for a generalized linear model (GLM), and which is better for high-volume data manipulation?

Description: Identifying and submitting all appropriate ICD-9 diagnosis codes to CMS will result in a higher risk score for the member and increased payments to Medicare Advantage plans. Milliman's HCC prevalence study illustrates the significant variations in HCC prevalence rates by age/gender, geography, and population type and will assist plans in benchmarking the effectiveness of their own diagnosis code capture efforts.

Description: Description of the vast variety of predictive analytics topics covered by the Forecasting & Futurism section and introduction to the July 2015 issue of the newsletter.

Description: An economic analysis of bariatric surgery and health care costs supports the case for health plans to cover these procedures.

Description: The following case study walks through the steps of a predictive modeling exercise to illustrate one way analytics can streamline the underwriting paradigm.

Description: Article discussing how geospatial imagery and high-resolution image technology are transforming property insurance.

Description: Analysis of possible mathematical relationships between hospital quality and hospital billed or allowed charges or other measures, using quality data as defined by the U.S. Department of Health and Human Services along with the Hospital Quality Alliance for their Hospital Quality Initiative.

Description: This article describes a method that can be useful when re-fitting a risk score model to a population of only moderate size. Learn how to use concepts like ridge regression to train a model with partial credibility.

Description: A study of disability experience examines cost burden and degree of disability, and how these measures vary across different subpopulations. This article summarizes a presentation given at the Society of Actuaries’ Living to 100: Survival to Advanced Ages International Symposium held on Jan. 7–9, 2008 in Orlando, Florida. An abstract of the author's paper was presented in the September 2008 issue of Long-Term Care News.

Description: The influenza pandemic of 1918-20 is recognized as having generally taken place in three waves, starting in the northern spring and summer of 1918. This pattern of three waves, however, was not universal: in some locations influenza seems to have persisted into or returned in 1920. The recorded statistics of influenza morbidity and mortality are likely to be a significant understatement. Limitations of these data can include nonregistration, missing records, misdiagnosis, and nonmedical certification, and may also vary greatly between locations. Further research has seen the consistent upward revision of the estimated global mortality of the pandemic, which a 1920s calculation put in the vicinity of 21.5 million. A 1991 paper revised the mortality as being in the range 24.7-39.3 million. This paper suggests that it was of the order of 50 million. However, it must be acknowledged that even this vast figure may be substantially lower than the real toll, perhaps as much as 100 percent understated.