EIA Revises Methodology for Historical Storage Data Comparisons
In order to sort out historical alignment issues in the rare year -- such as 2004 -- when there are 53 weekly natural gas storage reports instead of the usual 52 (see NGI, Jan. 10), the Energy Information Administration (EIA) said Thursday that beginning with the report for the week ended Jan. 7, it has implemented a daily interpolated average approach of calculating the year-ago and five-year average statistics.
"The daily interpolated average approach facilitates simple yearly comparisons of the data, unencumbered by possible confusion associated with ordering weekly data," the EIA said in its methodology appendix on computing the year-ago and five-year average for storage.
According to the agency, the calculations are based on daily linear interpolations of the working gas inventory values between each week reported in the Historical Weekly Storage Estimates Database. Using the historical data, daily data are derived by:
- Computing an average daily rate of stock change for each week based on the published weekly data; and
- Using this daily rate and the published weekly stock levels to estimate daily stock levels for each day in between the report days.
"With the resulting daily working gas time series, five-year averages are computed for each day of the year, using the preceding five years of daily estimates," the EIA said. "For example, the five-year average used as reference for 01-07-2005 includes the data for 01-07-2000, 01-07-2001, 01-07-2002, 01-07-2003, and 01-07-2004."
The change was greeted with widespread acceptance from natural gas futures market watchers, who found the old system incorrect and confusing.
"From what it appears, the EIA is now doing it the way it should be done," said Citigroup's Kyle Cooper. "It looks like they are going [by the] calendar. That's the way I do it and I think it seems to make the most sense."
Cooper added that the last system was designed by somebody who simply lined up the first Fridays of each year. "Then they realized that probably did not make any sense comparing a Jan. 1 to a Jan. 7," he said. "When they ran out of weeks last year, it really did not make a lot of sense. They adjusted it to make it more reasonable, which is obviously a good thing."
IFR Energy Services' Tim Evans was mostly positive on the EIA's methodology change. "The new system does appear to be more precise," Evans told NGI. "What I don't like is the fact that I haven't figured out the programming that I need to do yet to see in advance what they are going to use for the five-year comparison for this week's report. Eventually this will be easy, but I just haven't gotten there yet."
In an early estimate for this week's report, Cooper looks for a 100 Bcf withdraw. If realized, a takeaway of that size may be perceived by the market as bearish as it will pale in comparison to historical averages. According to the data released last Thursday by the EIA, the report to be released this Friday (delayed a day because of the inauguration) will be compared to a year-ago withdrawal of 155 Bcf and a five-year average draw of 133 Bcf. These data are available on the EIA's web site at: http://tonto.eia.doe.gov/oog/info/ngs/ngsstats.xls.
The EIA noted that Lower-48 states minimum and maximum values that are used in the storage graph included in the storage report are based on a similar treatment of the weekly data. However, regional working gas maxima or minima are calculated on a "noncoincident basis" and so will not necessarily sum to the maxima or minima determined for the Lower 48 states.