This preview shows pages 1–4. Sign up to view the full content.
Econometrics Homework 6
ARE 106
Fall 2008
A. Havenner
DESCRIPTION: Vostok Ice Core Data for 420,000 Years
Isotopes of Hydrogen and Oxygen have been used to develop Earth temperature
histories extending over 400,000 years.
The variable GT4 consists of time series data on global temperature
1
for the last
420,000 years as estimated from deep drilled cores at the Vostok station in Antarctica.
See, for example,
http://en.wikipedia.org/wiki/Vostok,_Antarctica
.
The dataset was
obtained from the National Oceanic and Atmospheric Administration and is available at
ftp://ftp.ncdc.noaa.gov/pub/data/paleo/icecore/antarctica/vostok/deutnat.txt
along with
their variable descriptions.
2
Here we are simply looking at temperature, and not its relation to CO
2
.
For a nicely
objective look at these data and their relation to CO
2
see the MIT student report at
http://web.mit.edu/angles2008/angles_Emmanuel_Quiroz.html
The beginning of the homework as assigned
The Gretl dataset Vostok.gdt has been emailed to you.
It contains time series data
on two variables, TIME and GT4.
The variable TIME is a technically adjusted version of
time, and we will use it as the time axis.
(Because the time axis is adjusted, the years
2001 to 5311 that Gretl assigns to a time series plot are meaningless and should only
be interpreted as arbitrary observation indexes.)
The variable GT4 is not identified.
While I normally will not run a regression without identifying the variables and much
more, in this case it may prove interesting and so I will ask you to do it.
1.
Use an XY scatter plot to plot TIME on the X axis and GT4 on the vertical
axis.
Print this plot and paste it into your homework.
Examine the plot:
what do you see?
1
The temperature measure is “Temperature difference wrt the mean recent time value” as reported.
Subtracting a
constant does not change the interpretation of the time coefficient so we can ignore this adjustment.
2
Note that NOAS’s text description of the columns does not match the column identification of the actual data.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document 0
[COMMENT "Cycles, with 5 peaks, of which the latest is declining after not reaching
prior maximums.
There may be a slight tilting upward, not because of the maximums
which are low, high, low, high, low, but rather because of the minimums, which slightly
tilt up.
The middle is indeterminate.
Objectively, there is nothing in this plot to suggest
that the last upswing is any different from the earlier ones (to my considerable
surprise).”]
2.
Run an Ordinary Least Squares (OLS) regression relating GT4 as the dependent
variable to a constant and TIME as the independent variables, and report the results in
standard format.
Model 1: OLS estimates using the 3311 observations 20015311
Dependent variable: GT4
coefficient
std. error
tratio
pvalue

const
4.63101
0.0846234
54.72
0.000
***
TIME
7.84995E07
4.98672E07
1.574
0.1155
Mean of dependent variable = 4.52391
Standard deviation of dep. var. = 2.89627
Sum of squared residuals = 27744.8
Standard error of the regression = 2.89563
Unadjusted Rsquared = 0.00075
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 03/31/2009 for the course ARE 106 taught by Professor Havenner during the Spring '09 term at UC Davis.
 Spring '09
 Havenner

Click to edit the document details