Prediction uncertainty of linear building energy use models with autocorrelated residuals

D. K. Ruch, J. K. Kissock, T Agami Reddy

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Autocorrelated residuals from regression models of building energy use present problems when attempting to estimate retrofit energy savings and the uncertainty of the savings. This paper discusses the causes of autocorrelation in energy use models and proposes a method to deal with autocorrelation. A hybrid of ordinary least squares (OLS) and autoregressive (AR) models is developed to accurately predict energy use and give reasonable uncertainty estimates. Only linear models are considered because both the data and the physical theory for many commercial buildings support this choice (Kissock, 1993). A procedure for model selection is presented and tested on data from three commercial buildings participating in the Texas LoanSTAR program. In every case examined, the hybrid OLS-AR model provided the best estimate of energy use and the most robust estimate of uncertainty.

Original languageEnglish (US)
Pages (from-to)63-68
Number of pages6
JournalJournal of Solar Energy Engineering, Transactions of the ASME
Volume121
Issue number1
StatePublished - Feb 1999
Externally publishedYes

Fingerprint

Autocorrelation
Uncertainty
Energy conservation

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Fuel Technology
  • Renewable Energy, Sustainability and the Environment
  • Mechanical Engineering

Cite this

Prediction uncertainty of linear building energy use models with autocorrelated residuals. / Ruch, D. K.; Kissock, J. K.; Reddy, T Agami.

In: Journal of Solar Energy Engineering, Transactions of the ASME, Vol. 121, No. 1, 02.1999, p. 63-68.

Research output: Contribution to journalArticle

@article{f42f82d53ed343a3a228aad9087e7f15,
title = "Prediction uncertainty of linear building energy use models with autocorrelated residuals",
abstract = "Autocorrelated residuals from regression models of building energy use present problems when attempting to estimate retrofit energy savings and the uncertainty of the savings. This paper discusses the causes of autocorrelation in energy use models and proposes a method to deal with autocorrelation. A hybrid of ordinary least squares (OLS) and autoregressive (AR) models is developed to accurately predict energy use and give reasonable uncertainty estimates. Only linear models are considered because both the data and the physical theory for many commercial buildings support this choice (Kissock, 1993). A procedure for model selection is presented and tested on data from three commercial buildings participating in the Texas LoanSTAR program. In every case examined, the hybrid OLS-AR model provided the best estimate of energy use and the most robust estimate of uncertainty.",
author = "Ruch, {D. K.} and Kissock, {J. K.} and Reddy, {T Agami}",
year = "1999",
month = "2",
language = "English (US)",
volume = "121",
pages = "63--68",
journal = "Journal of Solar Energy Engineering, Transactions of the ASME",
issn = "0199-6231",
publisher = "American Society of Mechanical Engineers(ASME)",
number = "1",

}

TY - JOUR

T1 - Prediction uncertainty of linear building energy use models with autocorrelated residuals

AU - Ruch, D. K.

AU - Kissock, J. K.

AU - Reddy, T Agami

PY - 1999/2

Y1 - 1999/2

N2 - Autocorrelated residuals from regression models of building energy use present problems when attempting to estimate retrofit energy savings and the uncertainty of the savings. This paper discusses the causes of autocorrelation in energy use models and proposes a method to deal with autocorrelation. A hybrid of ordinary least squares (OLS) and autoregressive (AR) models is developed to accurately predict energy use and give reasonable uncertainty estimates. Only linear models are considered because both the data and the physical theory for many commercial buildings support this choice (Kissock, 1993). A procedure for model selection is presented and tested on data from three commercial buildings participating in the Texas LoanSTAR program. In every case examined, the hybrid OLS-AR model provided the best estimate of energy use and the most robust estimate of uncertainty.

AB - Autocorrelated residuals from regression models of building energy use present problems when attempting to estimate retrofit energy savings and the uncertainty of the savings. This paper discusses the causes of autocorrelation in energy use models and proposes a method to deal with autocorrelation. A hybrid of ordinary least squares (OLS) and autoregressive (AR) models is developed to accurately predict energy use and give reasonable uncertainty estimates. Only linear models are considered because both the data and the physical theory for many commercial buildings support this choice (Kissock, 1993). A procedure for model selection is presented and tested on data from three commercial buildings participating in the Texas LoanSTAR program. In every case examined, the hybrid OLS-AR model provided the best estimate of energy use and the most robust estimate of uncertainty.

UR - http://www.scopus.com/inward/record.url?scp=0033077029&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033077029&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0033077029

VL - 121

SP - 63

EP - 68

JO - Journal of Solar Energy Engineering, Transactions of the ASME

JF - Journal of Solar Energy Engineering, Transactions of the ASME

SN - 0199-6231

IS - 1

ER -