ATMS 391: Homework 10
Problem 1
Perform a random sampling exercise on the Chicago August temperature dataset to determine the 90% confidence interval of the mean temperature for the month when randomly subsetting the data for 10, 50, 100, and 500 samples.
Plot your results using an errorbar plot: [http://matplotlib.org/1.2.1/examples/pylab_examples/errorbar_demo.html]
Problem 2
Using the global precipitation data file (prec.mon.mean.nc
), for January and July, fit a Gamma distribution to the monthly precipitation accumulations for the record of the data. Plot maps for January and July separately of the:
mean precipitation
standard deviation of precipitation
Gamma alpha parameter
Gamma beta parameter
Using a chi-squared test, and separating the data into 10 equally-spaced bins covering the range of values at each point, indicate on a map the locations where the fit is statistically significant at the 90% level. (i.e., see if you can or cannot reject the null hypothesis that the data follow a Gamma distribution as fitted).
Read this as it will illustrate the procedure. [http://www.stat.yale.edu/Courses/1997-98/101/chigf.htm]