Presenting “Estimating the Net Social Benefits of the NFIP” at the Benefit-Cost Society Conference in Februrary

The abstract is below. Conference information is available here.

The National Flood Insurance Program (NFIP) was created by Congress in 1968 to provide insurance and prevention against flood risk and to shift some rebuilding costs off the federal budget. The program, administered by FEMA, includes a flood mitigation grants component available to communities and a financial insurance component available to individuals and businesses. The program is self-supporting, while indebted to the federal government, and has been criticized for its environment and economic impacts. Multi-state flood events since 2000, such as Hurricanes Katrina (2005) and Sandy (2012) have stressed the NFIP’s finances forcing some to question the benefits and costs of the program.

Estimating the net social benefits through benefit-cost analysis requires an in-depth analysis of the insurance and grant components of the NFIP. This presentation will outline the development of a sufficient statistics for measuring the impacts of the NFIP’s insurance component. This sufficient statistic takes into account the premiums, claims payments, and potential losses for a policyholder and measures their impacts on society. This statistic is calculated using aggregate financial information provided by the program.

Combining the results of the insurance component sufficient statistic with estimates of the net social impacts of flood mitigation grants available from other sources, it is possible to obtain a first-order estimate of the net social benefits of the entire NFIP at the national level, both retrospectively and prospectively. In addition, it is possible to estimate the net social benefits of the NFIP at the state level and, through distributional weighting, obtain a second-order state level estimate of the distributionally weighted net social benefits of the NFIP. The framework can also provide higher-order estimates using disaggregated data sources.

Mercurial Histories

I put a lot of stuff into Mercurial including research, letters, and miscellaneous documents as they are being developed. When a new project shares some common element (such as, for instance, page formatting) with an older project, I’ll fork from the older project into the new one leaving them with a common history, even if they have no other logical or other connection between the two.

Simply Statistics: On weather forecasts, Nate Silver, and the politicization of statistical illiteracy

simplystatistics:

As you know, we have a thing for statistical literacy here at Simply Stats. So of course this column over at Politico got our attention (via Chris V. and others). The column is an attack on Nate Silver, who has a blog where he tries to predict the outcome of elections in the…

I’ve revised your code for efficiency:

# Set initial parameters
percentObama = 0.505
sdObama = 0.01
n = 1000

# Simulate n elections
simulatedPercentObama = rnorm(n,mean=percentObama,sd=sdObama)

# Calculate the percent of times Obama wins
percentObamaWin = mean(simulatedPercentObama > 0.5)
percentObamaWin