|
Activity Number:
|
413
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Business and Economic Statistics Section
|
| Abstract - #304030 |
|
Title:
|
Seasonality and Trends in the Temperature Anomaly Data from Goddard Institute for Space Studies
|
|
Author(s):
|
Peter B. Kenny*+ and Tucker S. McElroy
|
|
Companies:
|
PBK Research and U.S. Census Bureau
|
|
Address:
|
53 LIndal Crescent, Enfield, International, EN2 7RP, United Kingdom
|
|
Keywords:
|
Seasonality ; Trends ; Global warming
|
|
Abstract:
|
The GISS data show the differences between monthly average temperatures and the averages in the period 1951--1980; hence any stable seasonality should be eliminated. The published data for the entire globe, for the hemispheres and for 8 latitude bands are analyzed with X-12-ARIMA. It is found that there is identifiable seasonality in several of the series, particularly the whole globe, Northern hemisphere and latitudes 44N to 64N. The default trend chosen by X-12-ARIMA is rather short and looks implausible as a representation of the data. Using the longest possible Henderson (95 terms) a much more plausible trend is produced. The long-term trends for the series are very different, showing that global warming is far from uniform. The recent trend for some series moves downward; confidence limits for the latest trends are shown as a guide to whether this movement is likely to persist.
|
