This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Stat 133, Fall 2009 Homework 4: Spatial Data Due Thursday Nov. 5 at 11 pm on bSpace In this homework you will study the recent climate of California. The file CAtemp.RData contains data from 174 weather stations in California for the years 1960 through 2008. 1 The years 1961- 1990 are often used as a reference period for calculating the “typical” climate. This is because the amount of data available for this period is high relative to earlier years, and temperatures were not rising as rapidly as they have in recent years. Your task in this homework is to make maps of 1) the average California temperature from 1961-1990 and 2) the deviations from this average for each year from 2001-2008. To do this you will need to statistically interpolate the temperatures at the stations to a grid covering the state. 1. Average temperatures I’ve provided the R code to do the interpolation for the average temperatures from 1961- 1990 in the file HW4.R. Your first task is to step through this code and make sure you understand what it is doing. The main idea is that if we knew what the true covariance was between stations, we could use this to interpolate in an optimal way (ie, with the smallest variance possible). Since we don’t know the covariance, we first estimate it by calculating the covariogram and then fitting a function to it using nonlinear least squares. Then we plugthe covariogram and then fitting a function to it using nonlinear least squares....
View Full Document
This note was uploaded on 12/20/2009 for the course STAT 133 taught by Professor Staff during the Fall '08 term at University of California, Berkeley.
- Fall '08