NEMS Forecast Evaluation Methodology
(Draft)
This
is a working document prepared as a job of work (DE-AP01-06EI38129.A000) on
behalf of the Energy Information Administration (EIA) in order to solicit
advice and comment on methods for evaluating EIA forecasts.
Summary
The purpose of the methodology proposed here is to present a methodology for assessing the accuracy of National Energy Modeling System (NEMS) projections. NEMS is configured to project future energy product production and consumption in a fashion that accounts for the wide range of detailed circumstances that result in multi-market economic equilibria for energy products. In addition to market forces NEMS accounts for technological change and the impact of government actions and policies. The evaluation methodology presented here calls for constructing statistical approximations of important energy market relationships implicit to NEMS, e.g., residential and commercial sector energy consumption. The approximations will be derived by regression analyses of the NEMS solutions prepared in support of the Annual Energy Outlook (AEO); or, other special studies. The approximations are to be specified to account for important, explanatory relationships, e.g., the elasticity of sectoral energy consumption to the sectoral price of energy. Based on this, “differences” between NEMS projections and the actual values of the variables projected can be partitioned among general uncertainty, “errors” in projecting explanatory variables, structural changes in market behavior, and transitory influences such as the weather. An illustration of the approach is provided for residential and commercial sector demand for delivered energy, utilizing the versions of NEMS used in support of the 1998 – 2000 AEO’s.
November 2006
NEMS Forecast Evaluation Methodology
(DRAFT)
I. Background. In
principle, an evaluation of NEMS forecast accuracy that accounts for the
various sources of differences between forecast and actual values can be
readily accomplished. The form of the model, in brief, involves projecting a
variety of important influences on energy markets, e.g., technology and the
parameters of consumer behavior. Based on the projections of these “conditional
variables” the model then determines the corresponding multi-energy-market
equilibria with the associated projections of energy product prices and
quantities. When a forecast period actually occurs, and the actual values of
the conditional variables are known, the model can be re-run using the actual
values for model assumptions and the resulting projections compared to the
actual prices and quantities. The impact of individual conditional variables
can be determined by differentially using the original projections and the
actual values, ceteris paribus.
Routines within the model might also be amended to account for changes in
government policies, technologies, or consumer behavior. Extra-model influences
such as weather could also be accounted for in the historical period,
independent of NEMS, or using NEMS’ own
routines for processing the impact of weather variables. In general, the size
of the model and need to maintain, and then run, non-current versions of the model over many
years appear to make this approach impracticable. In any event, this approach
has not been attempted. Do do so, among other requirements, would entail
archiving and re-implementing versions of NEMS years and even decades after
their current use. A multitude of problems suggest that this approach is too
expensive, and perhaps not possible as a practical matter.
An alternative is to isolate the explanatory variables at issue for each model component to those that correspond to the basic economic forces associated with the energy markets represented by NEMS. Given this, the underlying interrelationships, e.g., price elasticities of supply and demand, can be determined for a current version of the model and then retained. In general, the method would be to selectively change important assumptions, ceteris paribus, and catalogue the corresponding sensitivities. Later, these sensitivities could be applied to actual data to determine the basis for forecast differences without having to archive and re-run the corresponding version of NEMS (This was done for RSTEM in Costello (2006) Reduced Form Energy Model Elasticities from EIA’s Regional Energy Model (RSTEM) , released 5/9/2006 as a one time report, for price and weather elasticities) A differential accounting for the impact of differences in projected versus actual “conditional variables” could still be approximated from the sensitivities. As before, extra-model, transitory influences on the historical values could also be accounted for. The basis for the approach is to detect the important market sensitivities implicit to NEMS’s representation of energy markets for each version of the model. This could be done for NEMS, although it would be, perhaps unreasonably, expensive to implement. Although the method proposed below is presented as an alternative, the actual conduct of a comparative statics analysis with NEMS components to verify model sensitivities should periodically be conducted.
II. Proposed Method. There is an alternative method for extracting the underlying sensitivities implicit to NEMS. The approach noted above would be to solve NEMS components, relative to a base or reference case, with each important assumption, e.g., the residential sector price of energy, changed, individually (with all other assumptions held constant), and compare results. The method proposed here as an alternative is to estimate the market sensitivities implicit to NEMS based upon the solutions prepared each year for the AEO. Among other ways, NEMS solutions are saved in a binary format and can be processed by the PC-based graphic interface Graf2000. Solution data have been saved in this format starting with those prepared for the 1998 AEO. This utility includes a regression component that enables regression analysis to be conducted using resident data; and, a data extraction routine that enables any collection of solution series to be extracted and input to other statistical procedures. Initially, the proposed method entails no additional resource requirements in terms of running NEMS or archiving versions of NEMS for use at a future time.1 Instead, the solution sets for the AEO versions of NEMS can be pooled for the projections to be evaluated at a future time.
The basic approach is to specify the underlying energy market supply and demand relationships in terms of their important, explanatory variables; and, given this, to estimate the relationships based upon NEMS solution data. The results of the estimates provide a description of the NEMS model version in terms of how energy markets are represented. The actual specifications utilized would be guided by the expertise of the EIA staff responsible for developing and maintaining individual model components. Since the solution sets can be readily archived, the actual regression analyses need not be conducted until the time that a model version is to be evaluated, although the outcomes of the regressions can have immediate diagnostic use in NEMS development.2 A demonstration of the general success of representing NEMS components via regression analysis is given in: Buck and Lady, “Approximation of Large, Computer-Based Economic Models,” presented at annual meetings of International Atlantic Economic Association on October 9, 2005 in New York City, New York. A copy of the paper can be downloaded from the link:
http://optima-com.com/buck_lady/AES_Paper.htm
It is proposed to configure the means of performing the regression analyses in a fashion that can be routinely conducted and maintained by EIA staff. The goal of the statistical analysis is to enable the errors in EIA forecasts to be explicitly decomposed with respect to influences such as the following:
Transitory Influences, e.g., weather, strikes, accidents, embargoes not accounted for in the projections.
Institutional Influences, e.g., changes in laws and regulations and changes in data series definitions compared to model assumptions.
Structural Influences, e.g., changes in resource availability or energy use technology compared to model assumptions.*
Errors in Projecting Conditional Variables, e.g., differences in the eventual values of activity drivers and other exogenous factors such as GDP and population.*
Errors in Behavioral Parameters, e.g., changes in consumer price sensitivities compared to those assumed by the forecasting methodology.*
Uncertainty, e.g., the residual error of
the projection method.*
The methodology for partitioning forecast differences among (such as) the influences outlined above is as follows for the items indicated by “*”, given the availability of actual data for previously forecast values.
The equations derived to represent the important relationships of supply and demand are re-run using the actual values for the explanatory variables. The actual values are substituted for the values used (or solved for) in the original projections, one explanatory variable at a time. This enables the identification of the influence of each explanatory variable separately. The equation is then re-run with all explanatory variables assigned their actual values. For this case, the residual error is due to general forecasting "uncertainty" or other, structural changes. Structural change is the issue of whether or not the values of the coefficients in the forecasting equation have changed for the forecast period compared to the model version to which the equation had been "fit." The equation is re-estimated and the results compared to the outcome of the estimation used to approximate the characteristics of the model. One method to assess if there were significant "differences" between the original, and revised, estimate is the Chow test (Chow, Gregory, "Tests of Equality Between Sets of Coefficients in Two Linear Regressions," Econometrica, 28, (July 1960), pp. 591-605.). Events influencing the actual data not accounted for by the forecasting equation will be identified and evaluated by EIA staff as appropriate.
NEMS design is intended to enable the differential assessment of changes in many influences on energy consumption and production. The present methodology for evaluating NEMS projections, e.g., as presented in:
http://www.eia.doe.gov/oiaf/analysispaper/forecasteval/index.html),
assesses the accuracy of projections
of seventeen aggregate variables, e.g., total energy consumption. A review of
this method was presented in: Winebrake, James and Denya Sakva (2006), "An
evaluation of errors in
III. An Example for Residential and Commercial
Sector Consumption of Delivered Energy. The example presented here is
intended to be illustrative and indicative of the methodology proposed, rather
than a definitive evaluation of NEMS projections of residential and commercial
sector energy demand.
To illustrate the proposed method, solution data from the 1998-2000 AEO versions of NEMS were assembled. A simple relationship was used to represent each sector’s demand:
Qt = A + B(Sector Price)t + C(Driver)t + DQt-1,
where Qt = sectoral consumption of delivered energy in year t, (Sector Price)t = the average price of energy delivered to the sector in year t ($ per million Btu), and (Driver)t = a sectoral “activity” variable. The driver for the residential sector was millions of households and the driver for the commercial sector was billions of square feet. The lagged endogenous variable picks up secular trends in such as energy intensity, the average size of a household, and other infrastructure or behavioral characteristics. Since the AEO hi/lo GDP cases represent shifts in energy demand, to accommodate issues of identification, the regressions were performed on data pooled from the AEO Base Case and the hi/lo WOP cases. The regressions were run on the solution data for the years 2000-2020. Generally, the fit of the regressions was exceptionally good (regression results are in Appendix C below). Below are the results for the residential sector. The regression equations were then used with “actual” data (the solution values in the AEO 2006 NEMS solutions) for projections for the years 2000-2005 and the differences compared to the original forecast values. An example is provided below for the projections of residential consumption of (delivered) energy in the year 2005 as projected by the 2000 AEO version of NEMS.
Model = N_ResAll_00 for Year = 20054
Residential Energy Consumption: NEMS - Actual
=-.28533 (as % =-2.44)
Source
NEMS Actual Impact Percent Elasticity Updated Elas
Price
13.02821 17.24 1.0614 9.08 -.277041 -.093418
Driver
111.3056 115.3573 -.2825 -2.42 .685312 .57847
Lag
11.33194 11.43521 -.016 -.14 .154078 .240553
HDD
4524 4228 .3802 3.25 .48155 n/a
CDD
1215 1444 -.2265 -1.94 .10487 n/a
Total
.9166 7.83
Uncertainty -1.2019 -10.28
The NEMS projection for 2005 was .28533 quads (or 2.44%) low as made for the 2000 AEO base case. The first five rows of the table evaluate the difference in the NEMS projection associated with actual, rather than assumed, values for each of the explanatory variables. The first three are from the regression equation while the assessment of the weather using heating degree days (HDD) and cooling degree days (CDD) was made independently (see Appendix A). In each row the value assumed for the 2000 AEO is compared with the actual value in 2005. The impact, in quads, is the difference in the projected versus actual value associated with the explanatory variable. For example the average price of energy assumed for the 2000 AEO was over 24% lower than the actual price in 2005.3 As a result, given the estimated residential sector price elasticity of -.277, the forecast value figured to be 1.0614 (9.08%) high. In contrast, the NEMS assumption for residential households was 3.5% low with an estimated impact (from the regression approximation of residential sector demand) of 2.42%. When the lag and weather effects were added in, the NEMS projection figured to be 7.83% high. But in fact it was 2.44% low. The corresponding uncertainty, or “unexplained” variation is, therefore, -10.28%.5
This differential analysis of forecast differences was performed for each of the years 2000-2005. A table summarizing the outcome of the analysis for each year is provided below.
Residential Sector Projection % Differences: Model =
N_ResAll_00
Year
2000 2001 2002 2003 2004 2005
NEMS%
-2.04 .91 -.84 -2.61 -.9 -2.44
Ajustments%
1.88 5.1 .84 .88 4.73 7.83
Uncertainty%
-3.92 -4.19 -1.67 -3.49 -5.64 -10.28
Inspection of the table reveals
that NEMS projections were generally low for each year (with 2001 the
exception) while the backcast adjustments suggested that NEMS projections
should have been high. A more detailed analysis would accompany a regression
equation with more explanatory variables. One hint for the example used is that
the “updated” price elasticity, derived by running the same regression equation
for the 2006 AEO version of NEMS shows a significantly lower price elasticity.
As given above, a formal analysis of structural change across NEMS versions
will be conducted using the Chow Test, or related statistical procedure. A plot
summarizing the comparison of the NEMS projection compared to the backcast
projection is provided below.

In the above plot “NEMS SIM” is the projection provided by the statistical approximation of the 2000 AEO NEMS version residential sector demand function. The “Backcast” is that provided by the regression equation using actual values for the explanatory variables. As noted, the projected values tended to be low, but the impacts of errors in projecting the corresponding explanatory variables led to the estimate that the NEMS projection should have been high, i.e., the backcast projections should be low. These results imply the significance of features of the NEMS residential sector demand relationship not accounted for in the regression equation utilized for this example; and/or, other, extra-NEMS impacts upon consumption not accounted for in the illustration developed here.
Appendix A presents a summary of the manner of accounting for the impact of weather on actual consumption. Appendix B presents the year by year workup and summary plot as above for each of the approximations of NEMS residential and commercial sector demand for the 1998-2000 versions of NEMS. Appendix C provides the regression results for the sectoral demand approximations.
Notes
1. A fair number of NEMS scenarios are run in support of each AEO, e.g., around forty were run for the AEO2006, in order to reveal important sensitivities and uncertainties. As the use of regression analysis of solution data is developed for the purposes proposed here, some number of additional runs might be formulated to facilitate the isolation of important influences for the ultimate evaluation of NEMS projections. As noted below, the formulation of weather scenarios might facilitate an accounting for weather effects as they cause actual variable values to deviate from forecast values.
2. Large changes in year to year implicit sensitivities could detect anomalies in model development.
3. The monetary units used for prices in the AEO lag the current period by two years. Accordingly, prices for the 2000 AEO are in $1998 while those in the 2006 AEO are in $2004. Actual prices for each AEO version are converted to the monetary units used in the contemporary AEO. The price index used to adjust monetary units to those used in each NEMS version was the “GDP Chain-Type Price Index,” as taken from Table B-3 (Appendix B) of the 2006 Economic Report of the President.
4. In the automated simulations of NEMS solutions supporting the examples presented here, “N_ResAll_##” stands for the regression-based approximation of NEMS residential sector energy demand with ## ranging from 98-00 for the 1998-2000 AEO version of NEMS. “N_ComAll_##” similarly stands for the approximation of NEMS commercial sector energy demand.
5. (Total Impact) + Uncertainty = NEMS – Actual.
Appendix A: Derivation of the Weather
Adjustment Impact Multiplier
Note: NEMS includes HDD and CDD explanatory variables in its
representation of sectoral demand. These variables are accounted for in a
somewhat different manner than used in RSTEM, the source of HDD and CDD
elasticities used here. The differences in the effects as estimated by both
methods has not been derived at the present time. Significantly, the petroleum
elasticity for the sectoral demand, as estimated from RSTEM for total demand,
appears to be low. As a result, the weather sensitivities used here may be
underestimates as applied to residential and commercial sector demand.
Background. The derivations below present a way to apply an
exogenous weather elasticity to actual data for weather and energy consumption
to compute the difference, in energy, between actual consumption and that level
of consumption that would have been the case for “normal” weather.
Let E = dQ/dW where E is the elasticity of energy demand (Q) to the condition of the weather as measured by heating degree days (HDD) or cooling degree days (CDD). In computing E let dQ and dW be changes in Q and the weather, i.e., HDD or CDD, as decimal proportions of the base quantities being used as the percentage bases for the elasticity, e.g., for the percentage change in Q = +10%, dQ = + .1. The analysis assumes that dW, i.e., dHDD or dCDD, and E are known. The problem, given this, is to compute dQ; and, given this, apply it to the data to derive the difference in consumption due to the weather in energy (quads).
Computing dW and dQ. Let,
dW = (Actual
Weather)/(
For example (AER Table 1.7) normal HDD = 4524 (based on 1971-2000 data) and actual HDD in the year 2000 were 4460. Given this,
dW = (4460/4524) – 1 = 98.585 – 1 = -.01415,
i.e., it was 1.415% warmer in 2000 as measured by HDD using “normal” HDD as the percentage base.
An estimate of the HDD elasticity of residential sector energy demand is .48155. Applying this estimate,
dQ/dW = E, dQ = E(dW) = .48155(-.01415) = -.006814 = Adjust.
This value is the estimate that residential sector demand in the year 2000 is .6814% lower due to a warmer fall/winter. Using normal weather demand as the percentage base,
Normal Q = (Actual Q)/(Adjust + 1),
where (Normal Q) is an estimate of consumption had normal weather been the case.
The Weather Impact Multiplier. Given the above, the difference between actual and “normal weather” demand is given by,
Impact = (Actual Q) –
(
Factoring out (Actual Q) gives,
Impact Multiplier = (1 – (1/(Adjust +1)).
For the example of residential energy demand in the year 2000,
Impact Multiplier = (1 – (1/(Adjust +1)) = (1 – (1/.99319)) = -.00686.
Actual residential sector demand in 2000 was 11.176 quads. From the above analysis, this value is .00686(11.176) = .0767quads low due to the warmer weather. Accordingly, a NEMS forecast that assumed normal weather would be .0767quads high. Below are the weather impact multipliers for residential and commercial sector energy demand for the years 2000-2005 as computed by the above method. The estimates of sectoral HDD and CDD elasticities are the weighted averages (using 2005 quantities) of the weather elasticities given in: http://www.eia.doe.gov/emeu/steo/pub/pdf/elasticities.pdf.
Weather Data
(AER Tables 1.7 and 1.8)
Year
HDD %
2000 4460 98.585 1229 101.152
2001 4223 93.347 1245 102.469
2002 4294 94.916 1393 114.65
2003 4460 98.585 1290 106.173
2004 4290 94.828 1232 101.399
2005 4228 93.457 1444 118.848
Normal 4524 1215
Residential in 2005
Fuel
Q(quads) Weight HDD Elas. CDD Elas.
Petroleum
1.585 .14094 .076 .016
Gas
5.08 .45172 .88 -.01
Electricity
4.581 .40734 .18 .263
Wt. Avg. Elas. .48155 .10487
Commercial in 2005
Fuel
Q(quads) Weight HDD Elas. CDD Elas.
Petroleum
.7899 .09564 .076 .016
Gas
3.103 .37572 .526 -.017
Electricity
4.366 .52864 .015 .11
Wt. Avg. Elas. .21283 .05329
Weather Impact Multipliers (Impact = Mult*(Actual
Q))
Residential
Commercial
Year
HDD CDD HDD CDD
2000 -.00686 .00121 -.00302 .00061
2001 -.0331 .00258 -.01436 .00131
2002 -.0251 .01513 -.01094 .00775
2003 -.00686 .00643 -.00302 .00328
2004 -.02554 .00147 -.01113 .00075
2005 -.03253 .01938 -.01412 .00994
Appendix B: Backcast Results And Workup

Residential
Sector Projection % Differences: Model = N_ResAll_98
Year 2000 2001 2002 2003 2004 2005
NEMS% 1.83 5.06 3.05 1.08 2.67 .92
Ajustments% 4.11 6.35 3.15 2.51 5.82 8.27
Uncertainty% -2.28 -1.3 -.1 -1.42 -3.15 -7.35
Model
= N_ResAll_98 for Year = 2000
Residential
Energy Consumption: NEMS - Actual = .20424 (as % = 1.83)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.45476 13.44 .2043 1.83 -.20489 -.093418
Driver 105.3386 105.189 .0075 .07 .475539 .57847
Lag 11.2109 10.66378 .1837 1.64 .333017 .240553
HDD 4524 4460 .0767 .69 .48155 n/a
CDD 1215 1229 -.0135 -.12 .10487 n/a
Total .4586
4.11
Uncertainty -.2544 -2.28
Model
= N_ResAll_98 for Year = 2001
Residential
Energy Consumption: NEMS - Actual = .55232 (as % = 5.06)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.46056 14.48 .4188 3.83 -.20489 -.093418
Driver 106.4929 109.0325 -.127 -1.16 .475539 .57847
Lag 11.38 11.17576 .0686 .63 .333017 .240553
HDD 4524 4223 .3616 3.31 .48155 n/a
CDD 1215 1245 -.0282 -.26 .10487 n/a
Total
.6938 6.35
Uncertainty -.1415 -1.3
Model
= N_ResAll_98 for Year = 2002
Residential
Energy Consumption: NEMS - Actual = .34203 (as % = 3.05)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.37254 13.33 .1986 1.77 -.20489 -.093418
Driver 107.6285 110.472 -.1422 -1.27 .475539 .57847
Lag 11.47643 10.92411 .1854 1.65 .333017 .240553
HDD 4524 4294 .2814 2.51 .48155 n/a
CDD 1215 1393 -.1696 -1.51 .10487 n/a
Total .3536 3.15
Uncertainty -.0116 -.1
Model
= N_ResAll_98 for Year = 2003
Residential
Energy Consumption: NEMS - Actual = .12474 (as % = 1.08)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.38588 13.98 .3306 2.87 -.20489 -.093418
Driver 108.7612 112.0135 -.1626 -1.41 .475539 .57847
Lag 11.55409 11.21206 .1148 1 .333017 .240553
HDD 4524 4460 .0789 .69 .48155 n/a
CDD 1215 1290 -.074 -.64 .10487 n/a
Total
.2877 2.51
Uncertainty -.163 -1.42
Model
= N_ResAll_98 for Year = 2004
Residential
Energy Consumption: NEMS - Actual = .30526 (as % = 2.67)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.30915 14.89 .5352 4.68
-.20489 -.093418
Driver 109.9131 113.6467 -.1867 -1.63 .475539 .57847
Lag 11.63279 11.50805 .0419 .37 .333017 .240553
HDD 4524 4290 .2921 2.55 .48155 n/a
CDD 1215 1232 -.0168 -.15 .10487 n/a
Total .6656 5.82
Uncertainty -.3603 -3.15
Model
= N_ResAll_98 for Year = 2005
Residential
Energy Consumption: NEMS - Actual = .10744 (as % = .92)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.31398 16.77 .9241 7.91 -.20489 -.093418
Driver 111.0799 115.3573 -.2139 -1.83 .475539 .57847
Lag 11.74047 11.43521 .1025 .88 .333017 .240553
HDD 4524 4228 .3802 3.25 .48155 n/a
CDD 1215 1444 -.2265 -1.94 .10487 n/a
Total
.9664 8.27
Uncertainty -.859 -7.35

Residential
Sector Projection % Differences: Model = N_ResAll_99
Year 2000 2001 2002 2003 2004 2005
NEMS% .77 3.88 1.82 -.42 .97 -.98
Ajustments% 3.82 6.25 3.17 2.36 5.35 8.41
Uncertainty% -3.05 -2.37 -1.35 -2.77 -4.37 -9.39
Model
= N_ResAll_99 for Year = 2000
Residential
Energy Consumption: NEMS - Actual = .08649 (as % = .77)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.97353 13.66 .1601 1.43 -.24661
-.093418
Driver 105.0142 105.189 -.0062 -.06 .344404 .57847
Lag 11.12915 10.66378 .2101 1.88 .447881 .240553
HDD 4524 4460 .0767 .69 .48155 n/a
CDD 1215 1229 -.0135 -.12 .10487 n/a
Total
.4271 3.82
Uncertainty -.3406 -3.05
Model
= N_ResAll_99 for Year = 2001
Residential
Energy Consumption: NEMS - Actual = .4236 (as % = 3.88)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.957 14.72 .4113 3.77 -.24661 -.093418
Driver 106.1805 109.0325 -.1011 -.93 .344404 .57847
Lag 11.26225 11.17576 .039 .36 .447881 .240553
HDD 4524 4223 .3616 3.31 .48155 n/a
CDD 1215 1245
-.0282 -.26 .10487 n/a
Total
.6826 6.25
Uncertainty -.259 -2.37
Model
= N_ResAll_99 for Year = 2002
Residential
Energy Consumption: NEMS - Actual = .20367 (as % = 1.82)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.85189 13.55 .1629 1.45 -.24661 -.093418
Driver 107.345 110.472 -.1109 -.99 .344404 .57847
Lag 11.34771 10.92411 .1912 1.71 .447881 .240553
HDD 4524 4294 .2814 2.51 .48155 n/a
CDD 1215 1393 -.1696 -1.51 .10487 n/a
Total
.355 3.17
Uncertainty -.1513 -1.35
Model
= N_ResAll_99 for Year = 2003
Residential
Energy Consumption: NEMS - Actual =-.04873 (as % =-.42)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.94609 14.22 .2972 2.58 -.24661 -.093418
Driver 108.525 112.0135 -.1237 -1.07 .344404 .57847
Lag 11.41573 11.21206 .0919 .8 .447881 .240553
HDD 4524 4460 .0789 .69 .48155 n/a
CDD 1215 1290 -.074 -.64 .10487 n/a
Total
.2703 2.36
Uncertainty -.319 -2.77
Model
= N_ResAll_99 for Year = 2004
Residential
Energy Consumption: NEMS - Actual = .11147 (as % = .97)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.00933 15.14 .4971 4.35 -.24661 -.093418
Driver 109.7295 113.6467 -.1389 -1.21 .344404 .57847
Lag 11.45932 11.50805 -.022 -.19 .447881 .240553
HDD 4524 4290 .2921 2.55 .48155 n/a
CDD 1215 1232 -.0168 -.15 .10487 n/a
Total
.6114 5.35
Uncertainty -.4999 -4.37
Model
= N_ResAll_99 for Year = 2005
Residential
Energy Consumption: NEMS - Actual =-.11463 (as % =-.98)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.04349 17.05 .9347 8 -.24661 -.093418
Driver 110.9643 115.3573 -.1558 -1.33 .344404 .57847
Lag 11.54668 11.43521 .0503 .43 .447881 .240553
HDD 4524 4228 .3802 3.25 .48155 n/a
CDD 1215 1444
-.2265 -1.94 .10487 n/a
Total
.9829 8.41
Uncertainty -1.0975 -9.39

Residential
Sector Projection % Differences: Model = N_ResAll_00
Year 2000 2001 2002 2003 2004 2005
NEMS% -2.04 .91 -.84 -2.61 -.9 -2.44
Ajustments% 1.88 5.1 .84 .88 4.73 7.83
Uncertainty% -3.92 -4.19 -1.67 -3.49 -5.64 -10.28
Model
= N_ResAll_00 for Year = 2000
Residential
Energy Consumption: NEMS - Actual =-.22746 (as % =-2.04)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.30314 13.81 .1277 1.14 -.277041 -.093418
Driver 105.3671 105.189 .0124 .11 .685312 .57847
Lag 10.70899 10.66378 .007 .06 .154078 .240553
HDD 4524 4460 .0767 .69 .48155 n/a
CDD 1215 1229 -.0135 -.12 .10487 n/a
Total
.2102 1.88
Uncertainty -.4377 -3.92
Model
= N_ResAll_00 for Year = 2001
Residential
Energy Consumption: NEMS - Actual = .099 (as % = .91)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.17413 14.89 .4324 3.96 -.277041 -.093418
Driver 106.539 109.0325 -.1738 -1.59 .685312 .57847
Lag 10.9483 11.17576 -.0353 -.32 .154078 .240553
HDD 4524 4223 .3616 3.31 .48155 n/a
CDD 1215 1245 -.0282 -.26 .10487 n/a
Total
.5567 5.1
Uncertainty -.4577 -4.19
Model
= N_ResAll_00 for Year = 2002
Residential
Energy Consumption: NEMS - Actual =-.09372 (as % =-.84)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.06647 13.7 .1597 1.42 -.277041 -.093418
Driver 107.7027 110.472 -.1931 -1.72 .685312 .57847
Lag 11.02311 10.92411 .0154 .14 .154078 .240553
HDD 4524 4294
.2814 2.51 .48155 n/a
CDD 1215 1393 -.1696 -1.51 .10487 n/a
Total
.0938 .84
Uncertainty -.1875 -1.67
Model
= N_ResAll_00 for Year = 2003
Residential
Energy Consumption: NEMS - Actual =-.30046 (as % =-2.61)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.06551 14.37 .3287 2.86 -.277041 -.093418
Driver 108.8805 112.0135 -.2184 -1.9 .685312 .57847
Lag 11.11834 11.21206 -.0146 -.13 .154078 .240553
HDD 4524 4460 .0789 .69 .48155 n/a
CDD 1215 1290 -.074 -.64 .10487 n/a
Total
.1006 .88
Uncertainty -.4011 -3.49
Model
= N_ResAll_00 for Year = 2004
Residential
Energy Consumption: NEMS - Actual =-.10327 (as % =-.9)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.08315 15.31 .5612 4.91 -.277041 -.093418
Driver 110.0828 113.6467 -.2485 -2.17 .685312 .57847
Lag 11.20759 11.50805 -.0467 -.41 .154078 .240553
HDD 4524 4290 .2921 2.55 .48155 n/a
CDD 1215 1232 -.0168 -.15 .10487 n/a
Total
.5412 4.73
Uncertainty -.6445 -5.64
Model
= N_ResAll_00 for Year = 2005
Residential
Energy Consumption: NEMS - Actual =-.28533 (as % =-2.44)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.02821 17.24 1.0614 9.08 -.277041 -.093418
Driver 111.3056 115.3573 -.2825 -2.42 .685312 .57847
Lag 11.33194 11.43521 -.016 -.14 .154078 .240553
HDD 4524 4228 .3802 3.25 .48155 n/a
CDD 1215 1444 -.2265 -1.94 .10487 n/a
Total
.9166 7.83
Uncertainty -1.2019 -10.28

Commercial
Sector Projection % Differences: Model = N_ComAll_98
Year 2000 2001 2002 2003 2004 2005
NEMS% -4.28 -2.97 -3.49 -3.59 -1.48 -2.7
Adjustments% 4.98 4.77 3.02 2.19 3.21 4.05
Uncertainty% -9.26 -7.74 -6.51 -5.78 -4.7 -6.74
Model
= N_ComAll_98 for year = 2000
Commercial
Energy Consumption: NEMS - Actual =-.348716 (as % =-4.28)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.3309 13.18 .0395 .49 -.063328 -.132454
Driver 74.89623 68.70575 .364 4.46 .565757 .318224
Lag 7.722946 7.76739 -.0168 -.21 .374241 .649403
HDD 4524 4460 .0246 .3 .21283 n/a
CDD 1215 1229 -.005 -.06 .05329 n/a
Total
.4063 4.98
Uncertainty -.755 -9.26
Model
= N_ComAll_98 for year = 2001
Commercial
Energy Consumption: NEMS - Actual =-.241118 (as % =-2.97)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.20863 14.43 .1035 1.27 -.063328 -.132454
Driver 75.71932 70.45488 .3095 3.81 .565757 .318224
Lag 7.803784 8.1525 -.1315 -1.62 .374241 .649403
HDD 4524 4223 .1166 1.44 .21283 n/a
CDD 1215 1245 -.0106 -.13 .05329 n/a
Total
.3875 4.77
Uncertainty -.6286 -7.74
Model
= N_ComAll_98 for year = 2002
Commercial
Energy Consumption: NEMS - Actual =-.287438 (as % =-3.49)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.1045 13.38 .0594 .72 -.063328 -.132454
Driver 76.51044 72.17918 .2547 3.09 .565757 .318224
Lag 7.880362 8.12148 -.0909 -1.1 .374241 .649403
HDD 4524 4294 .0902 1.09 .21283 n/a
CDD 1215 1393 -.0639 -.78 .05329 n/a
Total
.2495 3.02
Uncertainty -.5369 -6.51
Model
= N_ComAll_98 for year = 2003
Commercial
Energy Consumption: NEMS - Actual =-.299084 (as % =-3.59)
Source NEMS Actual Impact Percent Elasticity
Updated Elas
Price 12.05031 13.72 .0778 .93 -.063328 -.132454
Driver 77.33833 73.67056 .2157 2.59 .565757 .318224
Lag 7.957394 8.244832 -.1084 -1.3 .374241 .649403
HDD 4524 4460 .0252 .3 .21283 n/a
CDD 1215 1290 -.0273 -.33 .05329 n/a
Total
.1829 2.19
Uncertainty -.482 -5.78
Model
= N_ComAll_98 for year = 2004
Commercial
Energy Consumption: NEMS - Actual =-.122073 (as % =-1.48)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 11.94473 14.24 .1069 1.3 -.063328 -.132454
Driver 78.19029 75.03583 .1855 2.25 .565757 .318224
Lag 8.03783 8.336914 -.1128 -1.37 .374241 .649403
HDD 4524 4290 .0918 1.11 .21283 n/a
CDD 1215 1232 -.0062 -.08 .05329 n/a
Total
.2652 3.21
Uncertainty -.3873 -4.7
Model
= N_ComAll_98 for year = 2005
Commercial
Energy Consumption: NEMS - Actual =-.2273 (as % =-2.7)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 11.83853 15.85 .1868 2.22 -.063328 -.132454
Driver 79.01656 76.20448 .1654 1.96 .565757 .318224
Lag 8.122841 8.244914 -.046 -.55 .374241 .649403
HDD 4524 4228 .1191 1.41 .21283 n/a
CDD 1215 1444 -.0838 -.99 .05329 n/a
Total
.3415 4.05
Uncertainty -.5688 -6.74

Commercial
Sector Projection % Differences: Model = N_ComAll_99
Year 2000 2001 2002 2003 2004 2005
NEMS% -2.84 -1.41 -1.82 -2.02 0 -1.31
Adjustments% -1.72
-1.04 -3.27 -3.84 -2.57 .14
Uncertainty% -1.11 -.36 1.45 1.81 2.57 -1.45
Model
= N_ComAll_99 for year = 2000
Commercial
Energy Consumption: NEMS - Actual =-.231366 (as % =-2.84)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.57405 13.4 .104 1.28 -.172156 -.132454
Driver 62.89052 68.70575 -.2929 -3.59 .397353 .318224
Lag 7.820432 7.76739 .0287 .35 .536537 .649403
HDD 4524 4460 .0246 .3 .21283 n/a
CDD 1215 1229 -.005 -.06 .05329 n/a
Total
-.1406 -1.72
Uncertainty -.0908 -1.11
Model
= N_ComAll_99 for year = 2001
Commercial
Energy Consumption: NEMS - Actual =-.114683 (as % =-1.41)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.45059 14.67 .2794 3.44 -.172156 -.132454
Driver 63.5958 70.45488 -.3455 -4.25 .397353 .318224
Lag 7.921134 8.1525
-.1252 -1.54 .536537 .649403
HDD 4524 4223 .1166 1.44 .21283 n/a
CDD 1215 1245 -.0106 -.13 .05329 n/a
Total
-.0853 -1.04
Uncertainty -.0294 -.36
Model
= N_ComAll_99 for year = 2002
Commercial
Energy Consumption: NEMS - Actual =-.149878 (as % =-1.82)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.28535 13.6 .1655 2.01 -.172156 -.132454
Driver
64.25379 72.17918 -.3992 -4.84 .397353 .318224
Lag 8.006797 8.12148 -.0621 -.75 .536537 .649403
HDD 4524 4294 .0902 1.09 .21283 n/a
CDD 1215 1393 -.0639 -.78 .05329 n/a
Total
-.2695 -3.27
Uncertainty .1196 1.45
Model
= N_ComAll_99 for year = 2003
Commercial
Energy Consumption: NEMS - Actual =-.168493 (as % =-2.02)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.33296 13.95 .2036 2.44 -.172156 -.132454
Driver 64.93594 73.67056 -.4399 -5.28 .397353 .318224
Lag 8.094954 8.244832 -.0811 -.97 .536537 .649403
HDD 4524 4460 .0252 .3 .21283 n/a
CDD 1215 1290 -.0273 -.33
.05329 n/a
Total
-.3196 -3.84
Uncertainty .1511 1.81
Model
= N_ComAll_99 for year = 2004
Commercial
Energy Consumption: NEMS - Actual = .000088 (as % = 0)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.34807 14.48 .2684 3.26 -.172156 -.132454
Driver 65.61822 75.03583 -.4743 -5.75 .397353 .318224
Lag 8.168421 8.336914 -.0912 -1.11 .536537 .649403
HDD 4524 4290 .0918 1.11 .21283 n/a
CDD 1215 1232 -.0062 -.08 .05329 n/a
Total
-.2115 -2.57
Uncertainty .2116 2.57
Model
= N_ComAll_99 for year = 2005
Commercial
Energy Consumption: NEMS - Actual =-.110758 (as % =-1.31)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.33938 16.12 .476 5.64 -.172156 -.132454
Driver 66.28476 76.20448 -.4996 -5.92 .397353 .318224
Lag 8.245002 8.244914
0 0 .536537 .649403
HDD 4524 4228 .1191 1.41 .21283 n/a
CDD 1215 1444 -.0838 -.99 .05329 n/a
Total
.0117 .14
Uncertainty -.1225 -1.45

Commercial
Sector Projection % Differences: Model = N_ComAll_00
Year 2000 2001 2002 2003 2004 2005
NEMS% -4.16 -2.41 -2.67 -2.8 -.72 -1.8
Adjustments% -1.94 -2.49 -3.21 -3.71 -2.59 -.57
Uncertainty% -2.22 .09 .54 .9 1.87 -1.24
Model
= N_ComAll_00 for year = 2000
Commercial
Energy Consumption: NEMS - Actual =-.33897 (as % =-4.16)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 13.08415 13.55 .0281 .35 -.085136 -.132454
Driver 63.34077 68.70575 -.166 -2.04 .249884 .318224
Lag 7.710082 7.76739 -.0397 -.49 .686408 .649403
HDD 4524 4460 .0246 .3 .21283 n/a
CDD 1215 1229 -.005 -.06 .05329 n/a
Total
-.158 -1.94
Uncertainty -.181 -2.22
Model
= N_ComAll_00 for year = 2001
Commercial
Energy Consumption: NEMS - Actual =-.195572 (as % =-2.41)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.85606 14.83 .1192 1.47 -.085136 -.132454
Driver 64.20343 70.45488 -.1935 -2.38 .249884 .318224
Lag 7.81353 8.1525
-.2346 -2.89 .686408 .649403
HDD 4524 4223 .1166 1.44 .21283 n/a
CDD 1215 1245 -.0106 -.13 .05329 n/a
Total
-.2029 -2.49
Uncertainty .0073 .09
Model
= N_ComAll_00 for year = 2002
Commercial
Energy Consumption: NEMS - Actual =-.219833 (as % =-2.67)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.60261 13.75 .0693 .84 -.085136 -.132454
Driver 64.93169 72.17918 -.2243 -2.72 .249884 .318224
Lag 7.925908 8.12148 -.1354 -1.64 .686408 .649403
HDD 4524 4294 .0902 1.09 .21283 n/a
CDD 1215 1393 -.0639 -.78 .05329 n/a
Total
-.2641 -3.21
Uncertainty .0443 .54
Model
= N_ComAll_00 for year = 2003
Commercial
Energy Consumption: NEMS - Actual =-.233694 (as % =-2.8)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.52698 14.1 .095 1.14 -.085136 -.132454
Driver 65.62161 73.67056 -.2491 -2.99 .249884 .318224
Lag 8.024999 8.244832 -.1522 -1.83 .686408 .649403
HDD 4524 4460 .0252 .3 .21283 n/a
CDD 1215 1290 -.0273 -.33 .05329 n/a
Total
-.3085 -3.71
Uncertainty .0748 .9
Model
= N_ComAll_00 for year = 2004
Commercial
Energy Consumption: NEMS - Actual =-.05913 (as % =-.72)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.46017 14.64 .1316 1.6 -.085136 -.132454
Driver 66.34257 75.03583 -.269 -3.26 .249884 .318224
Lag 8.10322 8.336914 -.1618 -1.96 .686408 .649403
HDD 4524 4290 .0918 1.11 .21283 n/a
CDD 1215 1232 -.0062 -.08 .05329 n/a
Total
-.2136 -2.59
Uncertainty .1545 1.87
Model
= N_ComAll_00 for year = 2005
Commercial
Energy Consumption: NEMS - Actual =-.151769 (as % =-1.8)
Source NEMS Actual Impact Percent Elasticity Updated Elas
Price 12.3208 16.29 .2396 2.84 -.085136 -.132454
Driver 67.10787 76.20448 -.2815 -3.34 .249884 .318224
Lag 8.185784 8.244914
-.0409 -.49 .686408 .649403
HDD 4524 4228 .1191 1.41 .21283 n/a
CDD 1215 1444 -.0838 -.99 .05329 n/a
Total
-.0475 -.57
Uncertainty -.1043 -1.24
Appendix C: Regression Results
Endogenous
Variable:
Table
#2 Energy Consumption by Sector and Source (Quadrillion Btu per Year, Unless
Otherwise Noted)
Sector and Source: Residential: Delivered Energy...........
Exogenous
Variables:
# 1)
Table #3 Energy Prices by Sector and Source (1996 Dollars per Million Btu)
Sector and Source: Residential....................:
# 2)
Table #4 Residential Sector Key Indicators and Consumption (Quadrillion Btu per
year, Unless otherwise noted)
Key Indicators and Consumption: Households (millions): Total......................
# 3)
Lagged Table #2 Energy Consumption by Sector and Source (Quadrillion Btu per
Year, Unless Otherwise Noted)
Sector and Source: Residential: Delivered Energy...........
Exogenous
Variable Mean Coefficient Elasticity t-statistic
Variable#
1 12.16087 -.20739 -.20489 -5.242288
Variable#
2 117.0589 .050005 .475539 5.686485
Variable#
3 12.20893 .335754 .333017 2.896522
Constant 4.878567
Endogenous Mean SER R-sq LR-Multiplier
Variable 12.30925 .030533 .997571 1.505466348310
Data
pooled for the years 2000 to 2020 for the solutions given below:
aeo98b.ran hwop98.ran lwop98.ran
Endogenous
Variable:
Table
#2 Energy Consumption by Sector and Source (Quadrillion Btu per Year, Unless
Otherwise Noted)
Sector and Source: Residential: Delivered Energy...........
Exogenous
Variables:
# 1)
Table #3 Energy Prices by Sector and Source (1997 Dollars per Million Btu)
Sector and Source: Residential....................:
# 2)
Table #4 Residential Sector Key Indicators and Consumption (Quadrillion Btu per
year, Unless otherwise noted)
Key Indicators and Consumption: Households (millions): Total......................
# 3)
Lagged Table #2 Energy Consumption by Sector and Source (Quadrillion Btu per
Year, Unless Otherwise Noted)
Sector and Source: Residential: Delivered Energy...........
Exogenous
Variable Mean Coefficient Elasticity t-statistic
Variable#
1 12.78631 -.233292 -.24661 -6.478882
Variable#
2 117.4799 .03546 .344404 6.662139
Variable#
3 12.0023 .45137 .447881 5.39698
Constant 5.49542
Endogenous Mean SER R-sq LR-Multiplier
Variable 12.09579 .024701 .998236 1.822722053114
Data
pooled for the years 2000 to 2020 for the solutions given below:
aeo99b.ran hwop99.ran lwop99.ran
Endogenous
Variable:
Table
#2 Energy Consumption by Sector and Source (Quadrillion Btu per Year, Unless
Otherwise Noted)
Sector and Source: Residential: Delivered Energy...........
Exogenous
Variables:
# 1)
Table #3 Energy Prices by Sector and Source (1998 Dollars per Million Btu)
Sector and Source: Residential....................:
# 2)
Table #4 Residential Sector Key Indicators and Consumption (Quadrillion Btu per
year, Unless otherwise noted)
Key Indicators and Consumption: Households (millions): Total......................
# 3)
Lagged Table #2 Energy Consumption by Sector and Source (Quadrillion Btu per
Year, Unless Otherwise Noted)
Sector and Source: Residential: Delivered Energy...........
Exogenous
Variable Mean Coefficient Elasticity t-statistic
Variable#
1 13.06701 -.252014 -.277041 -4.669263
Variable#
2 116.8374 .069721 .685312 5.080185
Variable#
3 11.78577 .155396 .154078 .95813
Constant 5.202166
Endogenous Mean SER R-sq LR-Multiplier
Variable 11.88658 .031538 .997149 1.183986815122
Data
pooled for the years 2000 to 2020 for the solutions given below:
aeo2k.ran hwop2k.ran lwop2k.ran
Endogenous
Variable:
Table
#2 Energy Consumption by Sector and Source (quadrillion Btu, unless otherwise
noted)
Sector and Source: Residential: Delivered Energy
Exogenous
Variables:
# 1)
Table #3 Energy Prices by Sector and Source (2004 dollars per million Btu)
Sector and Source: Residential:
# 2)
Table #4 Residential Sector Key Indicators and Consumption (quadrillion Btu)
Key Indicators and Consumption: Households (millions): Total
# 3)
Lagged Table #2 Energy Consumption by Sector and Source (quadrillion Btu,
unless otherwise noted)
Sector and Source: Residential: Delivered Energy
Exogenous
Variable Mean Coefficient Elasticity t-statistic
Variable#
1 17.20498 -.066236 -.093418 -5.26068
Variable#
2 122.5325 .05759 .57847 5.694203
Variable#
3 12.07313 .243057 .240553 1.942199
Constant 3.347288
Endogenous Mean SER R-sq LR-Multiplier
Variable 12.19881 .110425 .977698 1.321103438435
Data
pooled for the years 2000 to 2020 for the solutions given below:
aeo2006.1119a.rlp2006.1201a.rahp2006.1130a.ra
Endogenous
Variable:
Table
#2 Energy Consumption by Sector and Source (Quadrillion Btu per Year, Unless
Otherwise Noted)
Sector and Source: Commercial: Delivered Energy...........
Exogenous
Variables:
# 1)
Table #3 Energy Prices by Sector and Source (1996 Dollars per Million Btu)
Sector and Source: Commercial.....................:
# 2)
Table #5 Commercial Sector Key Indicators and Consumption (Quadrillion Btu per
year, Unless otherwise noted)
Key Indicators and Consumption: Total Floorspace(bill. sq. ft.): Total......................
# 3)
Lagged Table #2 Energy Consumption by Sector and Source (Quadrillion Btu per
Year, Unless Otherwise Noted)
Sector and Source: Commercial: Delivered Energy...........
Exogenous
Variable Mean Coefficient Elasticity t-statistic
Variable#
1 11.60786 -.046574 -.063328 -12.408086
Variable#
2 82.13966 .0588 .565757 28.645436
Variable#
3 8.4741 .377014 .374241 18.558989
Constant 1.052853
Endogenous Mean SER R-sq LR-Multiplier
Variable 8.536895 .003931 .999913 1.605172507889
Data
pooled for the years 2000 to 2020 for the solutions given below:
aeo98b.ran hwop98.ran lwop98.ran
Endogenous
Variable:
Table
#2 Energy Consumption by Sector and Source (Quadrillion Btu per Year, Unless
Otherwise Noted)
Sector and Source: Commercial: Delivered Energy...........
Exogenous
Variables:
# 1)
Table #3 Energy Prices by Sector and Source (1997 Dollars per Million Btu)
Sector and Source: Commercial.....................:
# 2)
Table #5 Commercial Sector Key Indicators and Consumption (Quadrillion Btu per
year, Unless otherwise noted)
Key Indicators and Consumption: Total Floorspace(bill. sq. ft.): Total......................
# 3)
Lagged Table #2 Energy Consumption by Sector and Source (Quadrillion Btu per
Year, Unless Otherwise Noted)
Sector and Source: Commercial: Delivered Energy...........
Exogenous
Variable Mean Coefficient Elasticity t-statistic
Variable#
1 11.94725 -.125904 -.172156 -10.667412
Variable#
2 68.93118 .050367 .397353 14.026527
Variable#
3 8.663424 .541122 .536537 17.853183
Constant 2.081833
Endogenous Mean SER R-sq LR-Multiplier
Variable 8.737453 .008466 .999703 2.179228465953
Data
pooled for the years 2000 to 2020 for the solutions given below:
aeo99b.ran hwop99.ran lwop99.ran
Endogenous
Variable:
Table
#2 Energy Consumption by Sector and Source (Quadrillion Btu per Year, Unless
Otherwise Noted)
Sector and Source: Commercial: Delivered Energy...........
Exogenous
Variables:
# 1)
Table #3 Energy Prices by Sector and Source (1998 Dollars per Million Btu)
Sector and Source: Commercial.....................:
# 2)
Table #5 Commercial Sector Key Indicators and Consumption (Quadrillion Btu per
year, Unless otherwise noted)
Key Indicators and Consumption: Total Floorspace(bill. sq. ft.): Total......................
# 3)
Lagged Table #2 Energy Consumption by Sector and Source (Quadrillion Btu per
Year, Unless Otherwise Noted)
Sector and Source: Commercial: Delivered Energy...........
Exogenous
Variable Mean Coefficient Elasticity t-statistic
Variable#
1 12.22253 -.060376 -.085136 -10.431859
Variable#
2 69.99389 .030945 .249884 10.613828
Variable#
3 8.595922 .692154 .686408 30.985085
Constant 1.290166
Endogenous Mean SER R-sq LR-Multiplier
Variable 8.667882 .01009
.999555 3.248377435470
Data
pooled for the years 2000 to 2020 for the solutions given below:
aeo2k.ran hwop2k.ran lwop2k.ran
Endogenous
Variable:
Table
#2 Energy Consumption by Sector and Source (quadrillion Btu, unless otherwise
noted)
Sector and Source: Commercial: Delivered Energy
Exogenous
Variables:
# 1)
Table #3 Energy Prices by Sector and Source (2004 dollars per million Btu)
Sector and Source: Commercial:
# 2)
Table #5 Commercial Sector Key Indicators and Consumption (quadrillion Btu)
Key Indicators and Consumption: Total Floorspace (billion square Feet: Total
# 3)
Lagged Table #2 Energy Consumption by Sector and Source (quadrillion Btu,
unless otherwise noted)
Sector and Source: Commercial: Delivered Energy
Exogenous
Variable Mean Coefficient Elasticity t-statistic
Variable#
1 16.49366 -.073351 -.132454 -5.652417
Variable#
2 82.44836 .035254 .318224 4.778979
Variable#
3 8.997299 .659264 .649403 8.872552
Constant 1.505515
Endogenous Mean SER R-sq LR-Multiplier
Variable 9.133919 .085399 .990365 2.934823441021
Data
pooled for the years 2000 to 2020 for the solutions given below:
aeo2006.1119a.rlp2006.1201a.rahp2006.1130a.ra