With Bayesian analysis, we can fit any model; anything that we can do with frequentist or classical statistics, we can do with Bayesian statistics. In this next example, we will perform linear regression with both Bayesian inference and frequentist approaches. As we have covered the model creation and date parsing, we will go through things a little bit more quickly in this example. The data that we are going to use is the atmospheric CO2 over a span of about 1,000 years and the growth rate over the past 40 years, and then fit a linear function to the growth rate over the past 50-60 years.
The data for the last 50-60 years is from National Oceanic and Atmospheric Administration (NOAA) marine stations, surface sites. It can be found at http://www.esrl.noaa.gov/gmd/ccgg/trends/global.html , where you can download two datasets, growth rates, and annual means. The direct links to the data tables are ftp://aftp.cmdl.noaa.gov/products/trends...