Abstract
Omitted variables and measurement errors in explanatory variables frequently occur in hedonic price models. Ignoring these problems leads to biased estimators. In this paper, we develop a constrained autoregression–structural equation model (ASEM) to handle both types of problems. Standard panel data models to handle omitted variables bias are based on the assumption that the omitted variables are time-invariant. ASEM allows handling of both time-varying and time-invariant omitted variables by constrained autoregression. In the case of measurement error, standard approaches require additional external information which is usually difficult to obtain. ASEM exploits the fact that panel data are repeatedly measured which allows decomposing the variance of a variable into the true variance and the variance due to measurement error. We apply ASEM to estimate a hedonic housing model for urban Indonesia. To get insight into the consequences of measurement error and omitted variables, we compare the ASEM estimates with the outcomes of (1) a standard SEM, which does not account for omitted variables, (2) a constrained autoregression model, which does not account for measurement error, and (3) a fixed effects hedonic model, which ignores measurement error and time-varying omitted variables. The differences between the ASEM estimates and the outcomes of the three alternative approaches are substantial.
Similar content being viewed by others
Notes
A latent variable refers to a phenomenon that is supposed to exist but cannot be observed directly. Examples are welfare, quality of life, socioeconomic status. A latent variable is given empirical meaning by means of a correspondence statement or operational definition. Such a statement connects a latent variable with a set of observables. For instance, the latent variable socioeconomic status is operationalized (measured) by observed variables like income, education, and profession. For further details, see Folmer (1986) and the references therein.
Note that the spatial error model introduced by Cliff and Ord (1969) arises because of spatially correlated omitted variables. We furthermore refer to Anselin and Gracia (2008) and Kelejian and Prucha (2007) who present nonparametric approaches toward estimating covariance matrices affected by omitted variables. The approach presented in this paper is different in that it accounts for omitted variables in the regression equation and thus addresses both omitted variables bias of the estimator of the regression coefficients and of the covariance matrix of the estimators.
Since it is not needed for the remainder of this subsection, we suppress the index i.
An indicator may be related to more than one latent variable.
When the sample size increases, the asymptotic properties of the ML estimator start becoming effective and the impacts of deviation from normality start decreasing. Nevertheless, under non-normality, as reflected by among others the skewness and kurtosis of the data, and showing up in implausibly large standard errors, one may turn to robust standard error estimates (Jöreskog et al. 2000) or the bootstrap.
The composite variable represents the number of positive house attributes. Its coefficient is the average marginal price for an additional attribute, or improvement in one of the house materials.
Indicators can be categorized on the basis of the causal relationships to their latent constructs. A reflective indicator is the effect of a latent construct; a formative indicator is the cause (Bollen 1989, pp 64–65).
The IFLS is a longitudinal socioeconomic and health survey of Indonesian individuals and households. It was conducted by the RAND Institute (Strauss et al. 2004).
The data set relates to urban and rural residents. In this paper, we analyze the former only.
The effective sample size is the number of sample units with complete measurement, that is, without missing values.
Without the auxiliary autoregressions, all of the variances and covariances of a single indicator over time are used for identification of variances and covariances of its latent variable over time. By specifying the auxiliary autoregressions, the latent variables beyond the initial time period become endogenous and the parameters related to them are the autoregressive parameters and error model variances only. For instance, with three observations over time, there are three different variances and three different covariances of, say, House condition which can be used to identify six SEM parameters. For the auxiliary autoregression, however, only four of the six variances plus covariances are needed (i.e., two autoregressive parameters and two error term variances). Hence, there are two moments left that are available for identification of the time-invariant measurement error variance of the observed House Condition at the three time points.
The matrices of modification indices are available at http://blogs.unpad.ac.id/yusepsuparman/.
To economize on space, we do not present the estimates of the lagged coefficients. They are available at http://blogs.unpad.ac.id/yusepsuparman/.
The full set of modification indices can be obtained at http://blogs.unpad.ac.id/yusepsuparman/.
References
Anselin L, Gracia NL (2008) Error in variables and spatial effects in hedonic house price models of ambient air quality. Empir Econ 34(1):5–34
Anselin L, Gracia NL, Deichmann U, Lall S (2008) Valuing access to water: a spatial hedonic approach applied to Indian cities. Policy Res Working Paper, WPS4533, the World Bank
Arimah BC (1992) An empirical analysis of the demand for housing attributes in a third world city. Land Econ 68(4):366–379
Barrett LB, Waddell TE (1970) The cost of air pollution damages: a status report. Report, National Air Pollution Control Administration-Public Health Service-US Department of Health, Education and Welfare
Bentler PM (1989) EQS structural equations program manual. BMDP Statistical Software, Los Angeles
Bollen KA (1989) Structural equation with latent variables. Wiley, New York
Bollen KA, Brand JE (2010) A general panel model with random and fixed effects: a structural equations approach. Soc Forces 89(1):1–34
Browne MW, Cudeck R (1993) Alternative ways of assessing model fit. In: Bollen KA, Long JS (eds) Testing structural equation models. Sage, Newbury Park, pp 136–162
Butler RV (1982) The specification of hedonic indexes for urban housing. Land Econ 58(1):96–108
Chattopadhyay S (1999) Estimating the demand for air quality: new evidence based on the Chicago housing market. Land Econ 75(1):22–38
Chou CP, Bentler P, Satorra A (1991) Scaled test statistics and robust standard errors for non-normal data in covariance structure analysis: a monte carlo study. Br J Math and Stat Psychol 44(2):347–357
Clarke KA (2005) The phantom menace: omitted variable bias in econometric research. Confl Manag Peace Sci 22(4):341–352
Cliff AD, Ord JK (1969) The problem of spatial autocorrelation. In: Scott AJ (ed) London papers in regional sciences 1. Studies in Regional Science, Pion, London, pp 25–55
Cropper ML, Deck L, McConnell KE (1988) On the choice of functional forms for hedonic price functions. Rev Econ Stat 70(4):668–675
Diamantopoulos A, Riefler P, Roth KP (2008) Advancing formative measurement models. J Bus Res 61(12):1203–1218
Dinan TM, Miranowski JA (1989) Estimating the implicit price of energy efficiency improvement in the residential housing market: a hedonic approach. J Urban Econ 25(1):52–67
Dornbusch DM, Barrager SM (1973) Benefit of water pollution control on property values. Report, EPA-600/5-73-005, US Environmental Protection Agency
Engle RF, Lilien DM, Watson M (1985) A dynamic model of housing price determination. J Econ 28(3):307–326
Finch JF, West SG, MacKinnon DP (1997) Effects of sample size and nonnormality on the estimation of mediated effects in latent variable models. Struct Equ Modeling 4(2):87–105
Flores NE, Carson RT (1997) The relationship between the income elasticities of demand and willingness to pay. J Environ Econ Manag 33(3):287–295
Folmer H (1986) Regional economy policy. Measurement of its effects. Kluwer, Dordrecht
Folmer H, Oud JHL (2008) How to get rid of W: a latent variable approach to modeling spatially lagged variables. Environ Plan A 40(10):2526–2538
Fuller WA (1986) Measurement error models. Wiley, New York
Gamble HB, Langley CJ, Pashek RD Jr, Sauerlender OH, Twark RD, Downing RH (1973) Community effect of highways by property values. Report, the Federal Highway Administration-US Department of Transportation
Greene WH (2008) Econometric analysis. Prentice Hall, New Jersey
Gross DJ (1988) Estimating willingness to pay for housing characteristics: an application of the Ellickson bid-rent model. J Urban Econ 24(1):95–112
Gujarati DN (2004) Basic econometrics. McGraw-Hill, Singapore
Hausman JA, Taylor WE (1981) Panel data and unobserved individual effects. Econ 49(6):1377–1398
Hempel CG (1970) On the standard conception of scientific theories. In: Radner M, Winokur S (eds) Minnesota studies in the philosophy of science. University of Minnesota Press, Minneapolis
Jimenez E (1982) The value of squatter dwellings in developing countries. Econ Dev Cult Chang 30(4):739–752
Jöreskog KG (1973) A general method for estimating a linear structural equation system. In: Goldberger AS, Duncan OD (eds) Structural equation model in the social sciences. Freeman, San Fransisco, pp 1–56
Jöreskog K, Sörbom D (1996) LISREL 8: user’s reference guide. Scientific Software International, Chicago
Jöreskog K, Sörbom D, du Toit S, du Toit M (2000) LISREL 8: new statistical features. Scientific Software International, Chicago
Kelejian HH, Prucha IR (2007) HAC estimation in a spatial framework. J Econ 140(1):131–154
Kim CW, Phipps TT, Anselin L (2003) Measuring the benefit of air quality improvement: a spatial hedonic approach. J Environ Econ Manag 45(1):24–39
Little RJA (1988) A test of missing completely at random for multivariate data with missing values. J Am Stat Assoc 83(404):1198–1202
Little RJA, Rubin DB (1987) Statistical analysis with missing data. Wiley, New York
Malpezzi S (2008) Hedonic pricing models: a selective and applied review. In: O’Sullivan T, Gibb K (eds) Housing economics and public policy. Wiley, New York, pp 67–89
McMillen DP (2004) Airport expansions and property values: the case of Chicago O’Hare airport. J Urban Econ 55(3):627–640
Minguez R, Montero JM, Fernandex-Avilez G (2013) Measuring the impact of pollution on property prices in Madrid: objective versus subjective pollution indicators in spatial models. J Geogr Syst 15(2):169–191
Morrison DF (2004) Multivariate statistical methods. McGraw-Hill, New York
North J, Griffin C (1993) Water source as a housing characteristic: hedonic property valuation and willingness to pay for water. Water Resour Res 29(7):1923–1929
Nourse H (1967) The effect of air pollution on house value. Land Econ 43(1):181–189
Oud JHL, Delsing JMH (2010) Continuous time modeling of panel data by means SEM. In: van Montfort K, Oud JHL, Satorra A (eds) Longitudinal research with latent variables. Springer, Heidelberg, pp 201–244
Oud JHL, Folmer H (2008) A structural equation approach to models with spatially dependence. Geogr Anal 40(2):152–166
Ozanne L, Malpezzi S (1985) The efficacy of hedonic estimation with the annual housing survey: evidence from the demand experiment. J Econ Soc Meas 13(2):152–172
Pace RK, LeSage JP (2004) Spatial statistics and real estate. J Real Estate Financ Econ 29(2):147–148
Páez A (2009) Recent research in spatial real estate hedonic analysis. J Geogr Syst 11(4):311–316
Ridker RG, Henning JA (1967) The determinant of residential property value with special reference to air pollution. Rev Econ Stat 49(2):246–257
Rosen S (1974) Hedonic prices and implicit markets: product differentiation in price competition. J Polit Econ 82(1):34–55
Strauss J, Beegle K, Sikoki B, Dwiyanto A, Herawati Y, Witoelar F (2004) The third wave of the Indonesia Family Life Survey (IFLS3): overview and field report. Working paper, WR-144/1-NIA/NICHD, RAND
Tang J, Folmer H, Xue J (2013) Estimation of awareness and perception of water scarcity among farmers in the Guangzhong Plain, China, by mean of a structural equation model. J Environ Manag (forthcoming)
Tiwari P, Parikh J (1997) Demand for housing in the Bombay metropolitan region. J Policy Model 19(3):295–321
Tyrväinen L (1997) The amenity value of the urban forest: an application of the hedonic pricing method. Landsc Urban Plan 37(3):211–222
Verbeek M (2000) A guide to modern econometrics. Wiley, Chichester
Verbeek M, Nijman T (1992) Testing for selectivity bias in panel data model. Int Econ Rev 33(3):681–703
Waugh F (1928) Quality factor influencing vegetable price. J Farm Econ 10(2):185–196
Yusuf AA, Koundouri P (2004) Household valuation of domestic water in Indonesia: revisiting the supply driven approach. In: Koundouri P (ed) Econometrics informing natural resources management: selected empirical analyses. Edward Elgar, Cheltenham, pp 127–142
Yusuf AA, Koundouri P (2005) Willingness to pay for water and location bias in hedonic price analysis: evidence from the Indonesian housing market. Environ Dev Econ 10(3):821–836
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: Model specifications
1.1 FE
In SEM notation, the FE HP housing model reads as follows. For each wave, log(Rent Appraisal) is the only endogenous variable, while all the house characteristics are exogenous variables. Time-invariant unobserved heterogeneity is represented by the latent exogenous variable \( \left( {\xi_{10} } \right) \) which is correlated with the other exogenous variables and has a fixed unit regression coefficient in the three waves. The model does not account for measurement error, and hence, the relationships between the observed and the latent variables are identity relationships. The measurement models thus read:
\( {\varvec{\Uplambda}}_{y} = {\mathbf{I}}_{{\left( {3 \times 3} \right)}} ,\quad {\varvec{\Uplambda}}_{x} = \left[ {{\mathbf{I}}_{{\left( {9 \times 9} \right)}} {\mathbf{0}}_{\left( 9 \right)} } \right],\quad {\varvec{\tau}}_{y} = {\mathbf{0}}_{\left( 3 \right)} ,\quad {\varvec{\tau}}_{x} = {\mathbf{0}}_{\left( 9 \right)} ,\quad {\varvec{\Uptheta}}_{\varepsilon } = {\mathbf{0}}_{{\left( {3 \times 3} \right)}} \) and \( {\varvec{\Uptheta}}_{\delta } = {\mathbf{0}}_{{\left( {9 \times 9} \right)}} \).
The structural model parameter matrices are \( {\varvec{\alpha}}^{\prime } = \left[ {\begin{array}{*{20}c} {\alpha_{1} } & {\alpha_{2} } & {\alpha_{3} } \\ \end{array} } \right] \), \( {\mathbf{B}} = {\mathbf{0}}_{{\left( {3 \times 3} \right)}} \), \( {\varvec{\Upgamma}} = \left[ {\begin{array}{*{20}c} {\gamma_{11} } & {\gamma_{12} } & {\gamma_{13} } & 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & {\gamma_{24} } & {\gamma_{25} } & {\gamma_{26} } & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 & {\gamma_{37} } & {\gamma_{38} } & {\gamma_{39} } \\ \end{array} \, \begin{array}{*{20}c} 1 \\ 1 \\ 1 \\ \end{array} } \right] \), \( {\varvec{\Uppsi}} = diag\left[ {\begin{array}{*{20}c} {\psi_{11} } & {\psi_{22} } & {\psi_{33} } \\ \end{array} } \right] \), \( {\varvec{\kappa}}^{\prime } = \left[ {\begin{array}{*{20}c} {\kappa_{1} } & \cdots & {\kappa_{9} } & 0 \\ \end{array} } \right] \), and \( \Upphi = \left[ {\begin{array}{*{20}c} {\phi_{11} } & {} & {} & {} \\ {\phi_{21} } & {\phi_{22} } & {} & {} \\ \vdots & \vdots & \ddots & {} \\ {\phi_{10,1} } & {\phi_{10,2} } & \cdots & {\phi_{10,10} } \\ \end{array} } \right] \).
Under Constraint 2, γ 1k = 0.7892γ 2l and γ 2l = 1.1606γ 3m for \( \left( {k,l,m} \right) = \left\{ {\left( {1,4,7} \right),\left( {2,5,8} \right),\left( {3,6,9} \right)} \right\} \), while under the time-invariant coefficients assumption γ 1k = γ 2l and γ 2l = γ 3m for \( \left( {k,l,m} \right) = \left\{ {\left( {1,4,7} \right),\left( {2,5,8} \right),\left( {3,6,9} \right)} \right\} \).
1.2 SEM
The SEM HP structural model consists of the standard multiple regression model (2) in terms of latent variables, supplemented with the auxiliary autoregression models of the house characteristics for identification of the measurement error variances. The exogenous variables in this model are the house characteristics in wave-0. The exogenous and endogenous observed and latent variables are \( {\mathbf{x}}^{\prime } = \left[ {\begin{array}{*{20}c} {x_{1} } & {x_{2} } & {x_{3} } \\ \end{array} } \right] \), \( {\mathbf{y}}^{\prime } = \left[ {\begin{array}{*{20}c} {y_{1} } & \cdots & {y_{3} } \\ \end{array} } \right] \), \( {\varvec{\upxi}}^{\prime } = \left[ {\begin{array}{*{20}c} {\xi_{1} } & {\xi_{2} } & {\xi_{3} } \\ \end{array} } \right] \), \( {\varvec{\upeta}}^{\prime } = \left[ {\begin{array}{*{20}c} {\eta_{1} } & \cdots & {\eta_{9} } \\ \end{array} } \right] \). Note that the observed log(Rent Appraisal) variables are y 1, y 2, and y 6, because there is no lagged dependent variable in the structural model. The structural parameter matrices are \( {\mathbf{B}} = \left[ {\begin{array}{*{20}c} 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & {\beta_{23} } & {\beta_{24} } & {\beta_{25} } & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 & {\beta_{67} } & {\beta_{68} } & {\beta_{69} } \\ 0 & 0 & {\beta_{73} } & 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & {\beta_{84} } & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & {\beta_{95} } & 0 & 0 & 0 & 0 \\ \end{array} } \right] \) \( {\varvec{\Upgamma}} = \left[ {\begin{array}{*{20}c} {\gamma_{11} } & {\gamma_{12} } & {\gamma_{13} } \\ 0 & 0 & 0 \\ {\gamma_{31} } & 0 & 0 \\ 0 & {\gamma_{42} } & 0 \\ 0 & 0 & {\gamma_{53} } \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ \end{array} } \right] \), \( {\varvec{\alpha}} = \left[ {\begin{array}{*{20}c} {\alpha_{1} } \\ \vdots \\ {\alpha_{9} } \\ \end{array} } \right], \) \( {\varvec{\kappa}} = \left[ {\begin{array}{*{20}c} {\kappa_{1} } \\ {\kappa_{2} } \\ {\kappa_{3} } \\ \end{array} } \right] \) \( {\varvec{\Uppsi}} = diag\left[ {\begin{array}{*{20}c} {\phi_{11} } \\ \vdots \\ {\phi_{99} } \\ \end{array} } \right] \), and \( {\varvec{\Upphi}} = \left[ {\begin{array}{*{20}c} {\phi_{11} } & {} & {} \\ {\phi_{21} } & {\phi_{22} } & {} \\ {\phi_{31} } & {\phi_{32} } & {\phi_{33} } \\ \end{array} } \right] \).
Under Constraint 2, γ 1k = 0.7892β 2l and β 2l = 1.1604β 6m for \( \left( {k,l,m} \right) = \left\{ {\left( {1,3,7} \right),\left( {2,4,8} \right),\left( {3,5,9} \right)} \right\} \), while under time-invariant assumption γ 1k = β 2l and β 2l = β 6m for \( \left( {k,l,m} \right) = \left\{ {\left( {1,3,7} \right),\left( {2,4,8} \right),\left( {3,5,9} \right)} \right\} \).
The parameter matrices in the measurement models are \( {\varvec{\Uplambda}}_{y} = {\mathbf{I}}_{{\left( {9 \times 9} \right)}} , \) \( {\varvec{\Uplambda}}_{x} = {\mathbf{I}}_{{\left( {3 \times 3} \right)}} , \) \( {\varvec{\tau}}_{y} = {\mathbf{0}}_{\left( 9 \right)} \), \( {\varvec{\tau}}_{x} = {\mathbf{0}}_{\left( 3 \right)} \), \( {\varvec{\Uptheta}}_{\varepsilon } = diag\left[ {\begin{array}{*{20}c} 0 & 0 & {\theta_{33}^{\varepsilon } } & {\theta_{44}^{\varepsilon } } & {\theta_{55}^{\varepsilon } } & 0 & {\theta_{77}^{\varepsilon } } & {\theta_{88}^{\varepsilon } } & {\theta_{99}^{\varepsilon } } \\ \end{array} } \right] \), and \( {\varvec{\Uptheta}}_{\delta } = diag\left[ {\begin{array}{*{20}c} {\theta_{11}^{\delta } } & {\theta_{11}^{\delta } } & {\theta_{11}^{\delta } } \\ \end{array} } \right]. \) Constraint 1 is \( \theta_{kk}^{\delta } = \theta_{ll}^{\varepsilon } = \theta_{mm}^{\varepsilon } \) \( \left( {k,l,m} \right) = \left\{ {\left( {1,3,7} \right),\left( {2,4,8} \right),\left( {3,5,9} \right)} \right\} \).
1.3 AUT
The AUT model is (11). The endogenous variables in the model are log(Rent Appraisal) in wave-1 and wave-2, while all other variables are exogenous. The exogenous and endogenous observed and latent vectors are \( {\mathbf{x}}^{\prime } = \left[ {\begin{array}{*{20}c} {x_{1} } & \cdots & {x_{10} } \\ \end{array} } \right] \), \( {\mathbf{y}}^{\prime } = \left[ {\begin{array}{*{20}c} {y_{1} } & {y_{2} } \\ \end{array} } \right] \), \( {\varvec{\xi}}^{\prime } = \left[ {\begin{array}{*{20}c} {\xi_{1} } & \cdots & {\xi_{10} } \\ \end{array} } \right] \), and \( {\varvec{\eta}}^{\prime } = \left[ {\begin{array}{*{20}c} {\eta_{1} } & {\eta_{2} } \\ \end{array} } \right] \). Because of the absence of latent variables, the measurement models are \( {\varvec{\Uplambda}}_{y} = {\mathbf{I}}_{{\left( {2 \times 2} \right)}} \), \( {\varvec{\Uplambda}}_{x} = {\mathbf{I}}_{{\left( {10 \times 10} \right)}} \), \( {\varvec{\tau}}_{y} = {\mathbf{0}}_{\left( 2 \right)} \), \( {\varvec{\tau}}_{x} = {\mathbf{0}}_{{\left( {10} \right)}} \), \( {\varvec{\Uptheta}}_{\varepsilon } = {\mathbf{0}}_{{\left( {2 \times 2} \right)}} \), and \( {\varvec{\Uptheta}}_{\delta } = {\mathbf{0}}_{{\left( {10 \times 10} \right)}} \).
The structural model matrices are \( {\varvec{\alpha}} = \left[ {\begin{array}{*{20}c} {\alpha_{1} } \\ {\alpha_{1} } \\ \end{array} } \right] \), \( {\mathbf{B}} = \left[ {\begin{array}{*{20}c} 0 & 0 \\ {\beta_{21} } & 0 \\ \end{array} } \right] \), \( {\varvec{\Upgamma}} = \left[ {\begin{array}{*{20}c} {\gamma_{11} } & {\gamma_{12} } & {\gamma_{13} } & {\gamma_{14} } & {\gamma_{15} } & {\gamma_{16} } & {\gamma_{17} } & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & {\gamma_{25} } & {\gamma_{26} } & {\gamma_{27} } & {\gamma_{28} } & {\gamma_{29} } & {\gamma_{210} } \\ \end{array} } \right] \), \( {\varvec{\Uppsi}} = diag\left[ {\begin{array}{*{20}c} {\psi_{11} } & {\psi_{22} } \\ \end{array} } \right] \), \( {\mathbf{\kappa^{\prime}}} = \left[ {\begin{array}{*{20}c} {\kappa_{11} } & \cdots & {\kappa_{1010} } \\ \end{array} } \right] \), and \( {\varvec{\Upphi}} = \left[ {\begin{array}{*{20}c} {\phi_{11} } & {} & {} & {} \\ {\phi_{21} } & {\phi_{22} } & {} & {} \\ \vdots & \vdots & \ddots & {} \\ {\phi_{10,1} } & {\phi_{10,2} } & \cdots & {\phi_{10,10} } \\ \end{array} } \right] \).
Under Constraint 2, \( \gamma_{1k} = 1.1604\gamma_{2l} \) for \( \left( {k,l} \right) = \left\{ {\left( {5,8} \right),\left( {6,9} \right),\left( {7,10} \right)} \right\} \), γ 1k = −γ 11(0.7892γ 1l ) and γ 2l = − β 21 γ 1l for \( \left( {k,l} \right) = \left\{ {\left( {2,5} \right),\left( {3,6} \right),\left( {4,7} \right)} \right\} \). Under the time-invariant coefficients, assumption γ 1k = γ 2l for (k, l) = {(5, 8), (6, 9),(7, 10)}, γ 1k = − γ 11 γ 1l and γ 2l = − β 21 γ 1l for \( \left( {k,l} \right) = \left\{ {\left( {2,5} \right),\left( {3,6} \right),\left( {4,7} \right)} \right\} \).
Appendix 2: Time-invariant coefficients models
See Table 4.
Rights and permissions
About this article
Cite this article
Suparman, Y., Folmer, H. & Oud, J.H.L. Hedonic price models with omitted variables and measurement errors: a constrained autoregression–structural equation modeling approach with application to urban Indonesia. J Geogr Syst 16, 49–70 (2014). https://doi.org/10.1007/s10109-013-0186-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10109-013-0186-3
Keywords
- Hedonic housing price model
- Panel model
- Structural equation model
- Constrained autoregression
- Measurement error
- Omitted variable bias
- Urban Indonesia