Schedule for implementing WNGSR sample reselection

Monday, November 18, 2024
The Weekly Natural Gas Storage Report for the week ending November 8, 2024, will be the last report EIA publishes using the current sample. As previously announced, on Monday, November 18, 2024, at 3:00 p.m., EIA will revise estimates for the eight weeks covering September 20, 2024, through November 8, 2024, to gradually phase in the estimates from the new sample.

Thursday, November 21, 2024
Beginning with the WNGSR report published on November 21, 2024, containing data for the week ending November 14, 2024, published estimates will be based only on reports from operators in the new sample.

Why is EIA updating the Weekly Natural Gas Storage Report (WNGSR) sample now?

The WNGSR estimates of natural gas in underground storage are based on a sample survey, the EIA-912, Weekly Underground Natural Gas Storage Report. Sample reselections are necessary from time to time to capture changes in the survey population. EIA-191, Monthly Underground Natural Gas Storage Report, collects information from all storage operators, and it is the frame from which the WNGSR survey sample is drawn. Following standard statistical practice, we reselect survey samples periodically to more accurately reflect the population being surveyed.

Is EIA changing the established methodology to draw the sample or to estimate stocks and net changes?

The established methodology for drawing a new sample and estimating storage stocks and net changes has not changed. See Methodology for EIA Weekly Underground Natural Gas Storage Estimates for more information about the WNGSR sampling and estimation methodology.

When will the new 2024 sample replace the old 2018 sample?

The transition to estimates based on the new sample will occur in two steps. First, using the new sample, EIA will provide historical revisions to WNGSR estimates for the most recent eight weeks (September 20, 2024–November 8, 2024). We will publish these revisions on Monday, November 18, 2024, outside of the regular WNGSR publication schedule. Then, on Thursday, November 21, 2024, we will begin publishing estimates using the new sample in the WNGSR during the regular publication cycle.

How different is the sample size for the new 2024 sample compared with the old 2018 WNGSR sample?

The size of the new sample is very similar to the size of the old sample. The new sample includes 88 operators from a sampling frame of 133 operators in the Lower 48 states, and the old sample included 85 operators. The net effect of this sample reselection is that ten operators were added and seven operators were dropped.

How different is the coverage of the new 2024 sample compared with that for the old 2018 WNGSR sample?

The coverage of the new sample is very similar to the coverage of the old sample in terms of the measure of size that was used to select the sample, which was based on monthly working gas stocks collected in EIA-191. For the Lower 48 states, operators in the new sample account for 97% of the total measure of size approximately equaling the coverage for the old sample. Because of the relatively small number of operators (eight) in the Pacific region, we will base estimates of working natural gas stocks in that region on weekly reports from all of them, that is, a census data collection.

What is the difference in storage stock estimates between the new and old samples?

Since July 2024, differences in weekly estimates of working natural gas stocks between the new and old samples for the EIA-912 are not statistically significant at the 5% level of significance. Also, differences in interpolated monthly estimates derived from weekly stock estimates, compared with the monthly data based on the census survey information from the EIA-191, are similar for July 2024 and August 2024.

Will the WNGSR products, data files, or formats change when EIA begins using the new sample? Will there be opportunities to test any new products?

The WNGSR report tables, data files, and formats will not change with the new sample. Implementation of the new sample will not affect WNGSR products published on EIA’s website, including the wngsr.txt, wngsr.csv, and wngsr.json. No new weekly data files or tables will be produced. Therefore, we will not have a period of public testing.

What companies are currently being sampled? Can I get respondent-level data so I can build my own history?

The data submitted on the Form EIA-912 are protected under the Confidential Information Protection and Statistical Efficiency Act that precludes EIA from disclosing any identifiable information about the companies that report. However, monthly storage data, including operator name, location, and capacity, are available for all reporting operators in our public query system.

When will EIA transition to the new sample?

We will transition to the new sample starting with the report for the week ending November 15, 2024, which will be published on November 21, 2024. On November 18, 2024, we will publish revised estimates for the eight-week transition period (weeks ending September 20, 2024–November 8, 2024) that will incorporate a blending of estimates from the old and new samples. These revised estimates will follow the procedures for reporting revisions included in the WNGSR methodology.

Will EIA issue revised estimates as a result of the sample reselection?

On November 18, 2024, EIA will publish revised estimates for the eight-week transition period (September 20, 2024–November 8, 2024) that will incorporate a blending of new sample and old sample estimates. These revised estimates will follow the procedures for reporting revisions included in the WNGSR methodology.

How will EIA phase in the new sample?

The new sample will be phased in by revising previous published data over eight weeks to transition from the old (2018) sample to the new (2024) sample. We will calculate working natural gas estimates will be calculated as a weighted average of the estimates from the old sample and the new sample during the transition period. The published stock estimates will be calculated as a weighted average of the estimates based on both samples, with each ensuing week using a decreasing weight on the estimates from the old sample and an increasing weight on the estimates from the new sample (Table 1). We will release transition period estimates based on this weighted average ased as revisions to the WNGSR historical database on Monday, November 18. Revisions to estimated measures of sampling variability during the transition period also will be published at this time.

Table 1. Weights for combination of old (2018) and new (2024) samples
Week Ending	Weight on Estimated Stocks for 2018 Sample	Weight on Estimated Stocks for 2024 Sample
9/20/2024	7/8	1/8
9/27/2024	3/4	1/4
10/4/2024	5/8	3/8
10/11/2024	1/2	1/2
10/18/2024	3/8	5/8
10/25/2024	1/4	3/4
11/1/2024	1/8	7/8
11/8/2024	0	1

When was the last time EIA updated the sample?

EIA last updated the sample in September 2018.

Will EIA make other nonsampling changes to the WNGSR?

No, we are making no other changes to the data collection or data estimation methodologies. The last sample reselection in 2018 included a change in data collection practices for a few operators reporting working natural gas in the South Central region on the Form EIA-912 as well as the collection of a weekly census in the Pacific region. A discussion of these changes and an analysis of the effects of the 2018 sample reselection are available on our website.

Where can I find additional information about the new 2024 and old 2018 samples?

EIA provides additional information about the sample reselection for WNGSR.

For additional information, contact Jose Villar at jose.villar@eia.gov or (202) 586-9613.