Abstract
Accurate forecasts of home sales can provide valuable information for not only, policy makers, but also financial institutions and real estate professionals. Given this, our analysis compares the ability of two different versions of singular spectrum analysis (SSA) methods, namely recurrent SSA (RSSA) and vector SSA (VSSA), in univariate (UV) and multivariate (MV) frameworks, in forecasting seasonally unadjusted home sales for the aggregate US economy and its four census regions (Northeast, Midwest, South and West). We compare the performance of the SSA-based models with classical and Bayesian variants of the autoregressive (AR) and vector AR models. Using an out-of-sample period of 1979:8–2014:6, given an in-sample period of 1973:1–1979:7, we find that the UVVSSA is the best performing model for the aggregate US home sales, while the MV versions of the RSSA is the outright favorite in forecasting home sales for all the four census regions. Our results highlight the superiority of the nonparametric approach of the SSA, which in turn, allows us to handle any statistical process: linear or nonlinear, stationary or non-stationary, Gaussian or non-Gaussian.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
A simple Granger causality test between US home sales and seasonally unadjusted industrial production for the aggregate US economy revealed only one-way causality running from US home sales to US industrial production. This result, thus, supports the view of home sales serving as a leading indicator for economic activity, and also motivates our decision to leave out the use of fundamental information to forecast home sales. Complete details of the Granger causality tests are available upon request from the authors.
The optimal SSA code used in this study is available upon request.
It must be pointed out though, that the data on aggregate US home sales is available from 1963:1, but we start from 1973:1 for the sake of keeping our results comparable across models over the same in-sample and out-of-sample periods.
Based on the suggestion of an anonymous referee, we also estimated the AR and VAR models with 12 lags to account for the seasonal pattern in the data. However, the general conclusions of our results, i.e., the superiority of the SSA models (UVVSSA or MVRSSA), continued to hold. Complete details of these results are available upon request from the authors.
Note, we analyze five models for the US and nine for each of the four census regions. Given this, we would have \(5\times 5\) and \(9\times 9\) matrices of forecast comparison tests for each of the 12 horizons. Hence, we decided to take a parsimonious approach by comparing the two best model within the SSA and standard model categories.
References
Aye, G. C., Balcilar, M., Bosh, A., & Gupta, R. (2014). Housing and the business cycle in South Africa. Journal of Policy Modeling, 36(3), 471–491.
Bai, J., & Perron, P. (2003). Computation and analysis of multiple structural change models. Journal of Applied Econometrics, 18(1), 1–22.
Bauwens, L., Koop, G., Korobilis, D., & Rombouts, J. (2015). The contribution of structural break models to forecasting macroeconomic series. Journal of Applied Econometrics, 30, 596–620.
Doan, T. A., Litterman, R. B., & Sims, C. A. (1984). Forecasting and conditional projections using realistic prior distributions. Econometric Reviews, 3(1), 1–100.
Dua, P., & Miller, S. M. (1996). Forecasting Connecticut home sales in a BVAR framework using coincident and leading indexes. Journal of Real Estate Finance and Economics, 13(3), 219–235.
Dua, P., Miller, S. M., & Smyth, D. J. (1999). Using leading indicators to forecast US home sales in Bayesian VAR framework. Journal of Real Estate Finance and Economics, 18(2), 191–205.
Dua, P., & Smyth, D. J. (1995). Forecasting U.S. home sales using BVAR models and survey data on households’ buying attitudes for homes. Journal of Forecasting, 14(3), 167–180.
Golyandina, N., Nekrutkin, V., & Zhigljavsky, S. (2001). Analysis of time series structure: SSA and related techniques. New York: Chapman and Hall/CRC.
Gupta, R., Kabundi, A., Miller, S. M., & Uwilingiye, J. (2014). Using large data sets to forecast sectoral employment. Statistical Methods and Applications, 23(2), 229–264.
Gupta, R., Tipoy, C. K., & Das, S. (2010). Could we have predicted the recent downturn in home sales of the four US census regions? Journal of Housing Research, 19(2), 111–128.
Hassani, H. (2007). Singular spectrum analysis: Methodology and comparison. Journal of Data Science, 5(2), 239–257.
Hassani, H., Heravi, H., & Zhigljavsky, A. (2009). Forecasting European industrial production with singular spectrum analysis. International Journal of Forecasting, 25(1), 103–118.
Hassani, H., & Mahmoudvand, R. (2013). Multivariate singular spectrum analysis: A general view and new vector forecasting approach. International Journal of Energy and Statistics, 1(01), 55–83.
Hassani, H., Webster, A., Silva, E. S., & Heravi, S. (2015). Forecasting US tourist arrivals using optimal singular spectrum analysis. Tourism Management, 46, 322–335.
Khan, M. A. R., & Poskitt, D. S. (2011). Window length selection and signal–noise separation and reconstruction in singular spectrum analysis (No. 23/11). Monash University, Department of Econometrics and Business Statistics.
Leamer, E. E. (2007). Housing is the business cycle. Working paper 13428. National Bureau of Economic Research.
Litterman, R. B. (1981). A Bayesian procedure for forecasting with vector autoregressions. Working paper. Federal Reserve Bank of Minneapolis.
Litterman, R. B. (1986). Forecasting with Bayesian vector autoregressions—Five years of experience. Journal of Business and Economic Statistics, 4(1), 25–38.
McCracken, M. W. (2007). Asymptotics for out-of-sample tests of Granger causality. Journal of Econometrics, 140(2), 719–752.
Nyakabawo, W. V., Miller, S. M., Balcilar, M., Das, S., & Gupta, R. (2015). Temporal causality between house prices and output in the U.S.: A bootstrap rolling-window approach. The North American Journal of Economics and Finance, 33, 55–73.
Sims, C. A. (1980). Macroeconomics and reality. Econometrica, 48(1), 1–48.
Sims, C. A., Stock, J. H., & Watson, M. W. (1990). Inference in linear time series models with some unit roots. Econometrica, 58, 113–144.
Spencer, D. E. (1993). Developing a Bayesian vector autoregression model. International Journal of Forecasting, 9(3), 407–421.
Theil, H. (1971). Principles of econometrics. New York: Wiley.
Todd, R. M. (1984). Improving economic forecasting with Bayesian vector autoregression (pp. 18–29). Quarterly Review, Fall, Federal Reserve Bank of Minneapolis.
Tzagkarakis, G., Papadopouli, M., & Tsakalides, P. (2009). Trend forecasting based on Singular Spectrum Analysis of traffic workload in a large-scale wireless LAN. Performance Evaluation, 66(3), 173–190.
Acknowledgments
we would like to thank an anonymous referee for many helpful comments. However, any remaining errors are solely ours.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Hassani, H., Ghodsi, Z., Gupta, R. et al. Forecasting Home Sales in the Four Census Regions and the Aggregate US Economy Using Singular Spectrum Analysis. Comput Econ 49, 83–97 (2017). https://doi.org/10.1007/s10614-015-9548-x
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10614-015-9548-x
Keywords
- Home sales
- Forecasting
- Singular spectrum analysis
- Classical and Bayesian (vector) autoregressive models