case1502 {Sleuth2}

Global Warming

Description

The data are the temperatures (in degrees Celsius) averaged for the northern hemisphere over a full year, for years 1880 to 1987. The 108-year average temperature has been subtracted, so each observation is the temperature difference from the series average.

Usage

case1502

Format

A data frame with 108 observations on the following 2 variables.

Year

year in which yearly average temperature was computed, from 1880 to 1987

Temp

northern hemisphere temperature minus the 108-year average (degrees Celsius)

Source

Ramsey, F.L. and Schafer, D.W. (2002). The Statistical Sleuth: A Course in Methods of Data Analysis (2nd ed), Duxbury.

References

Jones, P.D. (1988). Hemispheric Surface Air Temperature Variations—Recent Trends Plus an Update to 1987, Journal of Climatology 1: 654–660.

Examples

str(case1502)

# Residuals from regression fit, ignoring autocorrelation
resids <- lm(Temp ~ Year, case1502)$res
# PACF plot shows evidence of 1st order auto correlation
acf(resids,type="partial")
# 1st autocorrelation coef.
acorr1 <- acf(resids,type="correlation",plot=FALSE)$acf[2] 

# Fit regression with filtered response and explanatory variables:
n <- length(case1502$Temp)
y <- with(case1502, Temp [2:n] - acorr1* Temp [1:(n-1)])
x <- with(case1502, Year [2:n] - acorr1* Year [1:(n-1)])
fit <- lm(y ~ x)
summary(fit)  # Interpret coefficient of x as coefficient of Year