Abstract
We introduce a semiparametric smooth coefficient estimator for recreation demand data that allows more flexible modeling of preference heterogeneity. We show that our sample of visitors each has an individual statistically significant price coefficient estimate leading to clearly nonparametric consumer surplus and willingness to pay (WTP) distributions. We also show mean WTP estimates that are different in economically meaningful ways for every demographic variable we have for our sample of beach visitors. This flexibility is valuable for future researchers who can include any variables of interest beyond the standard demographic variables we have included here. And the richer results, price elasticities, consumer surplus and WTP estimates, are valuable to planners and policymakers who can easily see how all these estimates vary with characteristics of the population of interest.
Similar content being viewed by others
Notes
Including the six observations has almost no impact on the estimation results in the current water quality scenario. For the improved water quality scenario, including them slightly changes the estimates of bandwidths, and the resulting mean WTP estimates are 18% higher. We return to this discussion in Sect. 5.
Silverman (1986) derives the optimal choice of bandwidth
$$\begin{aligned} h=\left( \frac{4{\hat{\sigma }}^5}{3n}\right) ^{\frac{1}{5}}\approx 1.06{\hat{\sigma }}n^{\frac{1}{5}},\end{aligned}$$where \({\hat{\sigma }}\) is the standard deviation of the samples. This approximation is called the normal distribution approximation or Gaussian approximation.
We first generate bootstrap error \(u_{i}^{*}\) based on the regression residual via Mammen’s two-point distribution. We then generate the bootstrap dependent variable \(y_{i}^{*}\) by adding the bootstrapped error to the fitted value from the regression, i.e. \(y_{i}^{*}=x_{i}{\hat{\beta }}(z_{i})+ u_{i}^{*}\). The bootstrap dependent variable \(y_{i}^{*}\) along with all other covariates are combined into a “bootstrap sample” which is used to estimate bootstrap coefficient \(\beta _{j}^{*}\). This process is repeated for a large number of times, 4000, and the standard deviation of all the bootstrap estimates of coefficient \(\beta _{j}\) is calculated as the standard error. The large number of draws was needed to insure convergence.
All the analysis in this paper is performed in R. The commands for estimation and testing are available in the ‘np’ package. Bootstrap and simulations are carried out using self-written codes.
To achieve stable results, a large number of simulations is needed; we used 4000 draws.
Following Bujosa et al. (2010), we also used the Wilcoxon non-parametric signed-rank test which produced results similar to the nonparametric equality of densities test. However, this test is usually used for observed outcomes, versus our estimated outcomes, so the results are not reported but available upon request.
For example, Ohio’s state gasoline tax is $0.28 per gallon while its eastern neighbor Pennsylvania has the nation’s highest state gasoline tax of $0.582, which is more than double.
The coefficient estimates are available upon request.
We use 500 Halton draws in the simulated maximum likelihood estimation.
References
Allenby GM, Arora N, Ginter JL (1998) On the heterogeneity of demand. J Mark Res 35(3):384–389. https://doi.org/10.2307/3152035
Allenby GM, Rossi PE (1999) Marketing models of consumer heterogeneity. J Econom 89(1–2):57–78. https://doi.org/10.1016/S0304-4076(98)00055-4
Andrews T (2001) A contingent valuation survey of improved water quality in the Brandywine River: an example of applied economics in the classroom. Pa Econ Rev 10:1–13
Awondo SN, Egan KJ, Dwyer DF (2011) Increasing beach recreation benefits by using wetlands to reduce contamination. Mar Resour Econ 26(1):1–15. https://doi.org/10.5950/0738-1360-26.1.1
Bach P, Farbmacher H, Spindle M (2017) Semiparametric count data modeling with an application to health service demand. Econom Stat. https://doi.org/10.1016/j.ecosta.2017.08.004
Baerenklau KA (2010) A latent class approach to modeling endogenous spatial sorting in Zonal recreation demand models. Land Econ 86(4):800–816. https://doi.org/10.3368/le.86.4.800
Boardman A, Greenberg D, Vining A, Weimer D (2017) Cost-benefit analysis: concepts and practice, th edn. Cambridge, Cambridge University Press
Bockstael NE, Strand IE Jr (1987) The effect of common sources of regression error on benefit estimates. Land Econ 63(1):11–20. https://doi.org/10.2307/3146652
Boxall P, Adamowicz W (2002) Understanding heterogeneous preferences in random utility models: a latent class approach. Environ Resour Econ 23(4):421–442. https://doi.org/10.1023/A:1021351721619
Bujosa A, Riera A, Hicks RL (2010) Combining discrete and continuous representations of preference heterogeneity: a latent class approach. Environ Resour Econ 47(4):477–493. https://doi.org/10.1007/s10640-010-9389-y
Cameron AC, Trivedi PK (2003) Essentials of count data regression. In: Baltagi BH (ed) A Companion to theoretical econometrics. Oxford, Blackwell Publishing Ltd, pp 331–348. https://doi.org/10.1002/9780470996249.ch16
Cameron AC, Trivedi PK (2005) Microeconometrics: methods and applications. Cambridge University Press, Cambridge
Cameron AC, Trivedi PK (2013) Regression analysis of count data. Cambridge University Press, Cambridge
Cooper J (2000) Nonparametric and semi-nonparametric recreational demand analysis. Am J Agric Econ 82:451–462. https://doi.org/10.1111/0002-9092.00038
Corrigan JR, Kling CL, Zhao J (2008) Willingness to pay and the cost of commitment: an empirical specification and test. Environ Resour Econ 40(2):285–298. https://doi.org/10.1007/s10640-007-9153-0
Creel M, Loomis J (1997) Semi-nonparametric distribution-free dichotomous choice contingent valuation. J Environ Econ Manag 32(3):341–358. https://doi.org/10.1006/jeem.1997.0972
Dillman D (2007) Mail and internet surveys, the tailored design method, 2nd edn. Wiley, New York
Egan K, Corrigan JR, Dwyer DF (2015) Three reasons to use annual payments in contingent valuation surveys: convergent validity, discount rates, and mental accounting. J Environ Econ Manag 72:123–136. https://doi.org/10.1016/j.jeem.2015.05.002
Egan K, Herriges J (2006) Multivariate count data regression models with individual panel data from an on-site sample. J Environ Econ Manag 52(2):567–581. https://doi.org/10.5950/0738-1360-26.1.1
Englin J, Shonkwiler JS (1995) Modeling recreation demand in the presence of unobservable travel costs: toward a travel price model. J Environ Econom Manag 29(3):368–377. https://doi.org/10.1006/jeem.1995.1053
Frölich MF (2006) Non-parametric regression for binary dependent variables. Econom J 9(3):511–540. https://doi.org/10.1111/j.1368-423X.2006.00196.x
Gurmu S, Rilstone P, Stern S (1999) Semiparametric estimation of count regression models. J Econom 88(1):123–150. https://doi.org/10.1016/S0304-4076(98)00026-8
Haab TC, McConnell KE (1996) Count data models and the problem of zeros in recreation demand analysis. Am J Agric Econ 78(1):89–102. https://doi.org/10.2307/1243781
Hall P, Li Q, Racine JS (2007) Nonparametric estimation of regression functions in the presence of irrelevant regressors. Rev Econ Stat 89(4):784–789. https://doi.org/10.1162/rest.89.4.784
Henderson DJ, Kumbhakar SC, Parmeter CF (2012) A simple method to visualize results in nonlinear regression models. Econ Lett 117(3):578–581. https://doi.org/10.1016/j.econlet.2012.07.040
Horowitz JL (1997) Bootstrap methods in econometrics: theory and numerical performance. In: Krepsand DM, Wallis KF (eds) Advances in economics and econometrics: theory and applications: seventh world congress, vol 3. Cambridge University Press, Cambridge, pp 188–222. https://doi.org/10.1017/CBO9781139051996.007
Horowitz JL (2001) The bootstrap. In: Heckmanand JJ, Leamer E (eds) Handbook of econometrics, vol 5. Elsevier, pp 3159–3228. https://doi.org/10.1016/S1573-4412(01)05005-X
Hynes S, Greene W (2013) A panel travel cost model accounting for endogenous stratification and truncation: a latent class approach. Land Econ 89(1):177–192. https://doi.org/10.3368/le.89.1.177
Jaime MM, Tudela AM (2011) Valuing a water recreation facility using semi parametric estimators in the travel cost method. Cuadernos de Economía 30:217–233
Landry CE, Liu H (2009) A semi-parametric estimator for revealed and stated preference data–an application to recreational beach visitation. J Environ Econ Manag 57(2):205–218. https://doi.org/10.1016/j.jeem.2008.05.002
Li Q, Huang CJ, Li D, Fu T-T (2002) Semiparametric smooth coefficient models. J Bus Econ Stat 20(3):412–422. https://doi.org/10.1198/073500102288618531
Li Q, Maasoumi E, Racine JS (2009) A nonparametric test for equality of distributions with mixed categorical and continuous data. J Econom 148(2):186–200. https://doi.org/10.1016/j.jeconom.2008.10.007
Li Q, Ouyang D, Racine JS (2013) Categorical semiparametric varying-coefficient models. J Appl Econom 28(4):551–579. https://doi.org/10.1002/jae.1261
Li Q, Racine JS (2007) Nonparametric econometrics: theory and practice. Princeton University Press, Princeton
Li Q, Racine JS (2010) Smooth varying-coefficient estimation and inference for qualitative and quantitative data. Econom Theory 26(06):1607–1637. https://doi.org/10.1017/S0266466609990739
Liu W (2014) Modeling gasoline demand in the United States: a flexible semiparametric approach. Energy Econ 45:244–253. https://doi.org/10.1016/j.eneco.2014.07.004
Liu W (2015) Gasoline taxes or efficiency standards? A heterogeneous household demand analysis. Energy Policy 80:54–64. https://doi.org/10.1016/j.enpol.2015.01.029
McLeod L (2011) A nonparametric versus latent class model of general practitioner utilization: evidence from Canada. J Health Econ 30(6):1261–1279. https://doi.org/10.1016/j.jhealeco.2011.08.005
Misra SK, Huang CL, Ott SL (1991) Consumer willingness to pay for pesticide-free fresh produce. West J Agric Econ 16(2):218–227
Morey E, Thacher J, Breffle W (2006) Using angler characteristics and attitudinal data to indentify environmental preference classes: a latent-class model. Environ Resour Econ 34(1):91–115. https://doi.org/10.1007/s10640-005-3794-7
Poe GL, Giraud KL, Loomis JB (2005) Computational methods for measuring the difference of empirical distributions. Am J Agric Econ 87(2):353–365
Poe GL, Severance-Lossin EK, Welsh MP (1994) Measuring the difference (X–Y) of simulated distributions: a convolutions approach. Am J Agric Econ 76(4):904–915. https://doi.org/10.2307/1243750
Poe GL, Welsh MP, Champ PA (1997) Measuring the difference in mean willingness to pay when dichotomous choice contingent valuation responses are not independent. Land Econ 73(2):255–267
Provencher B, Baerenklau KA, Bishop RC (2002) A finite mixture logit model of recreational angling with serially correlated random utility. Am J Agric Econ 84(4):1066–1075. https://doi.org/10.1111/1467-8276.00054
Racine JS (2008) Nonparametric econometrics: a primer. Found Trends Econom 3(1):1–88. https://doi.org/10.1561/0800000009
Robinson PM (1988) Root-N-consistent semiparametric regression. Econometrica 56(4):931–954. https://doi.org/10.2307/1912705
Scarpa R, Thiene M, Tempesta T (2007) Latent class count models of total visitation demand: days out hiking in the Eastern Alps. Environ Resour Econ 38(4):447–460. https://doi.org/10.1007/s10640-007-9087-6
Silverman BW (1986) Density estimation for statistics and data analysis. Chapman & Hall, London
Thompson E, Berger M, Blomquist G, Allen S (2002) Valuing the arts: a contingent valuation approach. J Cult Econ 26(2):87–113. https://doi.org/10.1023/A:1014426202110
Train KE (1998) Recreation demand models with taste differences over people. Land Econ 74(2):230–2394. https://doi.org/10.2307/3147053
Wedel M, Kamakura W, Arora N, Bemmaor A, Chiang J, Elrod T, Poulsen CS (1999) Discrete and continuous representations of unobserved heterogeneity in choice modeling. Mark Lett 10(3):219–232. https://doi.org/10.1023/A:100805431
Whitehead JC, Dumas CF, Herstine J, Hill J, Buerger B (2008) Valuing beach access and width with revealed and stated preference data. Marine Resour Econ 23(2):119–135. https://doi.org/10.1086/mre.23.2.42629607
Winkelmann R (2000) Econometric analysis of count data. Springer, Berlin
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Liu, W., Egan, K.J. A Semiparametric Smooth Coefficient Estimator for Recreation Demand. Environ Resource Econ 74, 1163–1187 (2019). https://doi.org/10.1007/s10640-019-00362-7
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10640-019-00362-7