Spatial Analysis of Tokyo Apartment Market

Size: px
Start display at page:

Download "Spatial Analysis of Tokyo Apartment Market"

Transcription

1 Spatial Analysis of Tokyo Apartment Market Morito Tsutsumi 1, Yasushi Yoshida, Hajime Seya 3, Yuichiro Kawaguchi 4 1 Department of Policy and Planning Sciences, University of Tsukuba Tennodai, Tsukuba-city, , Japan (tsutsumi@sk.tsukuba.ac.jp) Chiba University of Commerce Konodai, Ichikawa-city, 7-851, Japan (yyoshida@mug.biglobe.ne.jp) 3 Graduate School of Systems and Information Engineering, University of Tsukuba Tennodai, Tsukuba-city, , Japan (seya0@sk.tsukuba.ac.jp) 4 Graduate School of Finance, Accounting and Law, Waseda University Nihonbashi, Chuo-ku, Tokyo, , Japan (ykawaguchi@waseda.jp) Abstract. This study deals with apartment rent data which has been actually observed in Tokyo apartment market. It makes spatial analysis applying hedonic approach and discusses the spatial characteristics of the market. It also shows the significance of considering spatial autocorrelation of the errors in spatial hedonic model. Keywords: real estate, apartment rent, spatial autocorrelation, kriging, spatial autoregressive error model 1 Introduction Empirical researches on real estate data using spatial econometrics and spatial statistics approaches are developing. However, there are still quite a limited number of researches on Japanese market. Furthermore, most of them use socalled officially assessed land price data, which are provided by the Ministry of Land Infrastructure and Transport. The problem with their works is that the land price data are characterized by data distortions. More specifically, the data reflect ministry s opinion about land market and also contain appraisal smoothing bias. One of the advantages of this study is that it deals with the rent data which has been actually observed in Tokyo housing market. It employs hedonic approach for rent analysis. Although hedonic price or rent regression for real estate has been an indispensable tool in regional and transport analyses, not enough attention had been paid to its estimation in empirical applications except to multicollinearity of the variables. Spatial correlation and heteroscedasticity of error terms that violate the independence assumption on which the statistical analysis is based are often encountered and affect the parameter estimation, but little attention had been paid to them. In the next chapter, after brief description of data set, this study makes preliminary analysis and identifies the overall market characteristics. Then, it examines the residual spatial autocorrelation in a conventional regression model. It interprets the geographic distribution of the residuals from the view point of Tokyo s geological background. As a countermeasure for residual spatial autocorrelation, introducing sophisticated approaches developed in such as quantitative geography, spatial econometrics and spatial statistics have been used recently. However, introducing representative variable, such as dummy variable for zone, is a very simple idea and still often applied in practice. With regard to the latter, the study examines the effect of dummy variables specific to wards. Chapter 4 gives a brief overview of the methodology this study employs. It describes the models, more specifically, regression kriging developed in geostatistics and so-called spatial autoregressive error model in spatial econometrics which the study uses to consider residual spatial autocorrelation. In Chapter 4, the study estimates the models and compares those results and examines their performances. Furthermore, several parameter estimation methods are employed and their performances are compared. The results indicate the qualitative market characteristics. Then, the significance of spatial econometrics and spatial statistics approaches for the analysis of real estate data considering spatial dependence is discussed. Chapter 5 concludes the study. Preliminary Analysis based on Conventional Hedonic Model.1 Data Set and Overall Market Characteristics For empirical analysis, the data set about apartment rent and various kinds of attributes is provided by At Home Co., a provider of real estate market information and relevant support service agency in Japan. They have constructed a network for real estate information where about 56,000 real estate agencies are affiliates. The data set includes the data on the following attributes: position coordinate (latitude and longitude), rent, age, floor area, structure such as reinforced concrete, room type and number of rooms, sewage, gas, parking lot, air conditioner, bathroom, and so on. The present study uses a sample of 150 observations from the data. All of them are for January through May 006.

2 Morito Tsutsumi, Yasushi Yoshida, Hajime Seya, Yuichiro Kawaguchi Figure 1 shows the study area, 3 wards of Tokyo including three central wards of Chiyoda, Chuo and Minato. Chiyoda ward is the center of the city and in many ways the center of all Japan which has the Imperial Palace and the Diet, Tokyo Central Station and buildings of ministries and many large corporate headquarters. Chuo is historically the main commercial center of Tokyo, especially before World War II. In Minato ward, many embassies and many highclass apartments which are rented mainly by foreign executives of foreign firms are situated. Figure shows the location of observation properties and their rent values in terms of yen per square meter per month (ca. 1 US dollar = 10 Japanese yen, in May 007). The sample having the highest rent per square meter per month is located in Minato ward samples having high rent are located southwestward. In the Tokyo area, rents in the southern and western areas is said to be generally high. On the other hand, rent in the northern and eastern areas is said to be generally low. Tokyo s central three wards Chiyoda Chuo Minato Figure 1. Tokyo's 3 wards including central three wards Figure. Study area and location of observation points. Conventional Hedonic Model Let the following standard multiple linear regression model be as a basic model: y = X β + u (1) where y is an n 1 vector of log apartment rent; X is an n k matrix of the apartment attributes; β is an unknown parameter vector; and u is an n 1 vector of residuals. The standard assumptions we make about the residuals u in eq. (1) are E u = and Var( u) = σ I, where I is a unit matrix. As is well known, eq. (1) is referred to as hedonic model. ( ) 0 Table 1. Parameter estimates for the basic hedonic model Variable Coef. std dev. t constant train BoT, Tokyo * Odakyu * walk bus floor area (log.) age (sqr.) reinforced concrete * nos. of rooms one-room type * K-type * parking lot * self-locking * variance of error R-squared adjusted R-squared

3 Spatial Analysis of Tokyo Apartment Market 3 The explained variable is the logarithm of apartment rent [yen/ m per month]. The explanatory variables chosen after trial and error are: constant (intercept), time distance from the nearest station to central Tokyo by train or subway [minutes] (taking the average between the time required to Shinjuku station and the time to Tokyo or Otemachi station), Toei subway, which is operated by Bureau of Tokyo Metropolitan Government [dummy], Odakyu Electric Railway [dummy], time distance from the apartment to the nearest station by walk or bus [minutes], the logarithm of floor area [m ], the root square of age [year], reinforced concrete structure [dummy], the numbers of rooms, so-called one-room type where a very small kitchen is equipped [dummy], 1K type which is one room apartment with separate kitchen [dummy], parking lot [dummy], self-locking [dummy]. Parameter estimates for the conventional hedonic model given by Ordinary Least Squares (OLS) method are presented in Table 1..3 Detecting Spatial Pattern of the Residuals Figure 3 illustrates the spatial distribution of the residuals in the conventional hedonic model described in Section.. It indicates the existence of the residual spatial autocorrelation. More specifically, positive residuals are found southwest, especially high positive residuals in Minato ward. Negative residuals are found northeast. Figure 3. Spatial distribution of the residuals in conventional hedonic model Tests for the presence of spatial autocorrelation by Moran's statistic are carried out. The weighting system is defined as the functions of the physical distances between points. As the values of Moran's statistic depend on the assumed structure of W, in this study, the following five types of function are considered. i j j Table. Type of weight matrix component I II III IV V w = c d w = c d w = c exp( 0. 5 d ) w c exp d w = c exp d i = j w = 0 j j = ( ) Table 3 presents the result of Moran's tests for the residuals. In all cases, null hypothesis of no spatial dependence is rejected at the significance level of 1 %. Table 3. Test for spatial autocorrelation in the residuals of conventional hedonic model ii j j wight function I II III IV V moran's I Z Probabilities for normal distribution to exceed the value of Z

4 Morito Tsutsumi, Yasushi Yoshida, Hajime Seya, Yuichiro Kawaguchi.4 Introducing Dummy Variables against Residual Spatial Autocorrelation As a countermeasure for residual spatial autocorrelation, introducing representative variable, such as dummy variable for zone, is a very easy to handle and still often applied in practice although more sophisticated approaches have been used recently. Thus, the study examines the effect of dummy variables specific to zones. Chiyoda ward is chosen for the base to which no dummy variable is assigned. Consequently, dummy variables are added to the model. The result of parameter estimation is presented in Table 4. According to the t-values, 16 out of newly added dummy variables are estimated to be significant at the level of 5 %. Only those for Shibuya and Minato wards are positive and others negative. Table 4. Parameter estimates for the hedonic model having dummy variables for wards Dummy variables for wards Variable Coef. std. dev. t constant train BoT, Tokyo * Odakyu * walk bus floor area (log) age (sqr) reinforced concrete * nos. of rooms one-room type * K-type * parking lot * self-locking * Chuo * Minato * Shinjuku * Bunkyo * Taito * Sumida * Koutou * Shinagawa * Meguro * Ota * Setagaya* Shibuya * Nakano * Suginami * Toshima * Kita * Arakawa * Itabashi * Nerima * Adachi * Katsushika * Edogawa * variance of error R-squared adjusted R-squared Figure 4 and 5 illustrates the geographical distribution of the coefficients and t-values of dummy variables for wards. It apparently shows that the dummy variable for Shibuya ward, which is located west of Minato ward, is most significant. Shibuya is known as one of the fashion centers of Japan, particularly for young people. In addition, it includes well-known commercial and residential districts. On the other hand, significantly negative values are found north and east. These results correspond to those in the previous section and indicate the market structure.

5 Shibuya Minato Figure 4. Visualizing the coefficients of dummy variables for wards Figure 5. Visualizing the t-values for dummy variables Figure 6 illustrates the spatial distribution of the residuals in the present model. Compared with Figure 3, it does not indicate the existence of the residual spatial autocorrelation at a glance. Figure 6. Spatial distribution of the residuals after introducing dummy variables for wards Table 5 presents the result of Moran's tests for the residuals. In case of using type I and II of weight function, null hypothesis of no spatial dependence is again rejected at the significance level of 5 %. However, in case of using exponential types of weight function (type III to V), null hypothesis of no spatial dependence is not rejected. Although this result does not advocate dismantlement of spatial dependence in the residuals, it implies that introduction of such representative variables as widely used in practice let a model satisfy the assumption for OLS. However, this approach may increase the number of parameters, which lead to generalization problem, i.e., worsen the prediction ability. Table 5. Detection of spatial autocorrelation in the hedonic model with dummy variables for wards Weight Matrix I II III IV V moran's I Z p Thus, the study makes models without using dummy variables for wards in the following chapters.

6 Morito Tsutsumi, Yasushi Yoshida, Hajime Seya, Yuichiro Kawaguchi 3 Spatial Hedonic Modeling 3.1 Overview of Modeling Approach Spatial model is an essential tool for both practitioners and academics to analyze real estate data where spatial association or correlation is a key concept. Therefore, modeling techniques have been introduced from the fields of geography, geostatistics, spatial econometrics and so on (e.g. Benirschka (1994), Can (199), Dubin (1988), Velante et al. (005)). However, empirical analyses on real estate data are often faced with data problems. This study makes models based on both geostatistics and spatial econometrics. 3. Geostatistics Approach An alternative to consider the correlation among the residuals is the following regression kriging model: () I y = X β + ε, Σ Var ε = σ H ( φ) + τ, () where Σ is the n n covariance matrix of the errors, σ is so-called partial-sill variance, and τ is nugget variance. H (φ) is an n n matrix having ( i, j) th entry exp( si s j φ), where s i denotes the location coordinates of i th observation t and φ is a parameter. Let θ = ( β, σ, τ, φ). Velante et al. (005) uses the same type of model and demonstrates its significance by analyzing the asking rent data in US. They separate the results into three groups by the relative success of the spatial model over the conventional regression model. Kim et al. (003) uses kriging to estimate the values of unobserved explanatory variables for spatial hedonic model based on conventional spatial econometrics approach. On the contrary, this study uses regression kriging to estimate the rent values themselves similar to Valente et al (005). Knight et al. (1998) discusses a variety of data problems that confront real estate researches such as missing data, measurement error and censored data. They demonstrate impressive gains by using the Gibbs sampler to deal with missing data in a hedonic house price model. However, the deficiency with Kight et al. s work is that their models lack empirical spatial considerations. Tsutsumi et al. (006) employs Bayesian spatial modeling and improves upon the results without spatial consideration significantly. The present study employs the following four methods for parameter estimation. (i) Generalized least squares method and weighted least squares method (GLS-WLS): The method of weighted least squares is applied for fitting variogram model. Then the covariance matrix Σ is calculated based on the results and the trend parameter is estimated by generalized least squares method. (ii) Maximum likelihood method (ML): It assumes that the error obeys normal distribution: y = X β + ε, ε~ N( 0, σ H ( φ) + τ I), (3) then, maximize the log likelihood, L ( β,, τ y, X ) φ. (iii) Restricted maximum likelihood method (RML): According to Kitanidis and Lane (1985), maximum likelihood method often yields biased estimates especially unless the number of observation is large. Restricted maximum likelihood method is a counter measure for the problem and this method is also used in the present study. For more detail, see Kitanidis and Lane (1985). (iv) Markov Chain Monte Carlo (MCMC): We assume the prior distribution of θ, p(θ), can be formulated as p( θ ) = p( β ) p( σ ) p( τ ) p( φ ). Eq.() can be recast as a hierarchical model y θ, w ~ N( Xβ + w, τ I), where w σ, φ ~ N( 0, σ H ( φ)). The priors chosen for the present study are β ~ N( c, T ), σ ~ IG( aσ, bσ ), τ ~ IG( aτ, bτ ), and p ( φ) 1 φ, where IG(, ) denotes Inverse Gamma distribution. In this paper, we abbreviate to show full conditional distributions for sampling under these specifications due to space limitation. See Banerjee et al. (003) for more detail. As sampling methods, Gibbs sampler is used for ( β, w, τ, σ ) and Metropolis-Hastings algorithm for (φ).

7 3.3 Spatial Econometrics Approach Spatial Analysis of Tokyo Apartment Market 7 Another alternative to consider the correlation among the residuals is the following regression model called spatial autoregressive error model: y = X β + ε, ε = λ Wε + u, (4) where W = { w } is called spatial weight matrix, which denotes the effect of each zone, λ is a parameter, u is an error vector. In order to determine the effects of spatial autocorrelation, we must design the spatial weight matrix W. The weighting system is often defined as the functions of the physical distances between points such as shown in Table, where c j is a constant which leads w. (5) i =1 This study employs five types of weight matrix shown in Table. The study assumes that the error vector, u, obeys normal distribution u~n( 0, σ u I) and applies maximum likelihood (ML) method to estimate the parameters of the model. Kriging, explained in the previous section, assumes spatial stationarity which enables us to predict the value at arbitrary point or site. However, since spatial econometrics approach does not assume spatial stationarity, it cannot be used for predicting the value of the data which does not enter into parameter estimation. Haining (1990) calls the approach which develops a description for the observed data first and uses this to predict the values of unobserved data "sequential approach". Predicting based on sequential approach violates the assumptions on spatial weight matrix, so it is not suitable for the models based on spatial econometrics approach. An alternative is called "simultaneous approach" where unobserved value is regarded as missing value and both parameter values and missing values are estimated simultaneously. Referring to Martin (1984) and Haining (1990), suppose we have n samples and values on rent are missing from h of these. Let y = y o y p where y o denotes the ( n h) dimensional vector of observed rent values and y p denotes the 1 h dimensional vector of unknown rent values. Herewith, the covariance matrix of ε ( = ( I λ W ) u), Var ( ε), can be formulated as Var() ε V oo Vop (6) = Vpo Vpp so that V is the covariance matrix for the sub-vectors of y and y above mentioned. Maximizing log likelihood qr ( ) function, L β, σ, λ, y y, leads to the estimator of y, ŷ : u p o 1 ( Vˆ ) ( y X ˆ β ) p y ˆ = X ˆ β + Vˆ. (7) p p po oo o o p For more detail, see Martin (1984) or Haining (1990). To examine the prediction ability of the model, this study applies simultaneous approach and presents the results in the next chapter. q r 4 Empirical Results 4.1 Comparison among the Models and the Estimation Methods The estimation results of the basic regression model and the present spatial models, i.e., regression kriging and spatial autoregressive error model, are shown in Table 6 and 7. The number of iterations for MCMC is 10, 000 and the discarded (i.e. burn-in) is 1,000. As mentioned above, the present study tests five types of weight matrix for spatial autoregressive error model. These tables show significant success of the spatial models over the conventional regression model. It is interesting that kriging works better that spatial autoregressive error model despite of its unrealistic assumption, stationarity.

8 Morito Tsutsumi, Yasushi Yoshida, Hajime Seya, Yuichiro Kawaguchi Table 6. Comparative reuslt of parameter estimates for basic regression model and univesal kriging models Basic Regression Model Regression Kriging Estimation Method OLS WLS-GLS ML REML MCMC Variable Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t constant train BoT, Tokyo * Odakyu * walk bus floor area (log) age (sqr) reinforced concrete * nos. of rooms one-room type * K-type * parking lot * self-locking * variance of error nugget (tau^) partial-sill (sigma^) range (phi) R-squared adjusted R-squared * dummy variable Table 7. Comparative reuslt of parameter estimates for spatial autoregressive error model Weight Matrix I II III IV V Variable Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t constant train BoT, Tokyo * Odakyu * walk bus floor area (log) age (sqr) reinforced concrete * nos. of rooms one-room type * K-type * parking lot * self-locking * variance of error lambda R-squared adjusted R-squared loglikelihood Further Discussion In Table 6, estimated nugget plus partial sill by those methods except for REML equals c.a. 0.03, which corresponds with the variance of error estimated by OLS. This could imply the regression kriging model is identified properly. The values of estimated range which represents the extent of existing spatial correlation are shown in terms of kilometer. Thus, the estimated range is about 3 km to 10 km. Figure 7 illustrated the plot of semi-variance values and best-fitted semivariogram for WLS-GLS and ML. As is often pointed out, WLS-GLS leads to robust estimation, and consequently gives rather small range estimates in this case. From the view point range estimation, MCMC gives similar result. To identify the market characteristics and consider business strategy, these results are quite interesting. In order to examine the prediction ability of the models, we use 50 observations for validation apart from the 150 observations shown in Figure 1 and used for estimates, and calculate root mean square error (RMSE). The conventional regression model estimated by OLS and regression kriging models use the parameters presented in Table 6 to predict the rent values of 50 points. As mentioned in Section 3.5, spatial econometrics models require simultaneous approach to predict the values of the data which are not used for parameter estimation. The results based on the approach are presented in Table 8. Some estimates such as for the variable of one-room type drastically changes from those in Table 7. This study also applies simultaneous approach for kriging employing maximum likelihood method and presents the result in Table 9, which is similar to that in Table 6.

9 Spatial Analysis of Tokyo Apartment Market 9 ML WLS-GLS Figure 7. Plot of semi-variance values and best-fitted semivariogram Table 10 summarizes the results showing the calculated RMSE for these models. Among the four parameter estimation methods for regression kriging, ML gives the best root mean square error (RMSE). Nevertheless, other methods also work out. Spatial autoregressive models using the exponential types of weight function also works out. As is well known, the specification of spatial weight matrix affected the results to a remarkable degree. Table 8. Resutl of parameter estimates for spatial autoregression error model based on simltanoues approach Weight I II III IV V Variable Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t Coef. std. dev. t constant train BoT, Tokyo * Odakyu * walk bus floor area (log) age (sqr) reinforced concrete * nos. of rooms one-room type * K-type * parking lot * self-locking * variance of error lambda loglikelihood Table 9. Resutl of parameter estimates for kriging by maximum likelihood based on simltanoues approach Variable Coef. std. dev. t constant train BoT, Tokyo * Odakyu * walk bus floor area (log) age (sqr) reinforced concrete * nos. of rooms one-room type * K-type * parking lot * self-locking * nugget partial-sill 0.06 range 8.070

10 Morito Tsutsumi, Yasushi Yoshida, Hajime Seya, Yuichiro Kawaguchi Table 10. Summary of the prediction results for the models Approach sequential simultaneous non-spatial spatial Model Regression Kriging Spatial Autoregressive Error Model OLS Estimation Method WLS ML REML MCMC ML ML Weigh Matrix I II II IV V RMSE Concluding Remarks This study analyzed the Tokyo's 3 wards apartment market applying hedonic approach. Preliminary analysis gave the overall market characteristics and suggested the spatial consideration in the error terms. Then, the study applied regression kriging developed in geostatistics and so-called spatial autoregressive error model in spatial econometrics. Below are main conclusions. The residuals in the conventional hedonic model indicated strong spatial association in the market. As a countermeasure for residual spatial autocorrelation, the study applied the way of introducing dummy variables for wards as representative variables, which is still often used in practice, and verified its effect. The results showed the advantage of regression kriging over conventional regression model in prediction precision although it assumes stationarity across the area, which appears to be unrealistic. The study applied several parameter estimation methods for kriging and presented that the estimates of range particularly differs among the methods. The results based on kriging identified the range which represents the extent of existing spatial correlation. The estimated range is about 3 km to 10 km. The results also showed the advantage of spatial autoregressive error model over conventional regression model. The study presented that the simultaneous approach for spatial autoregressive error model in which the unobserved rent is regarded as missing value and both observed and missing values are contained, made good predictions. The study applied several types of spatial weight matrix for spatial autoregressive error model and compared the results. As is often mentioned, the specification of spatial weight matrix affected the results to a remarkable degree. Acknowledgement. The authors would like to thank At Home Co., Ltd. for their collaboration on providing the apartment rent data. References 1. Anselin, L. (1988) Spatial Econometrics: Methods and Models. Dordrecht : Kluwer Academic.. Banerjee, S., Carlin, B.P. and Gelfand, A.E.(003): Hierarchical Modeling and Analysis for Spatial Data, Chapman & Hall/CRC. 3. Benirschka, M. and Binkley, J. K.(1994) : Land price volatility in a geographically dispersed market, Journal of the American Journal of Agricultural Economics, 76, Can, A.(199): Specific and estimation of hedonic housing price models, Regional Science and Urban Economics,, Cressie, N.(1985): Fitting variogram models by weighted least squares. Mathematical Geology, 17, Dubin, R.A. (1988): Estimation of regression coefficient in the presence of spatially autocorrelated error terms. The Review of Economics and Statistics, 70, Haining, R.(1990): Spatial Data Analysis in the Social and Environmental Sciences, Cambridge University Press. 8. Kim C. W., Phipps T. T. and Anselin L. (003): Measuring the benefits of air quality improvement: a spatial hedonic approach, Journal of Environmental Economics and Management, 45 (1), Kitanidis,P.K. and Lane,R.W.(1985): Maximum Likelihood Parameter Estimation of Spatial Processes by the Gauss-Newton Method, Journal of Hydrology,79, Knight, J. R., Sirmans, C. F., Gelfand, A. E. and Ghosh S. K.(1998): Analyzing Real Estate Data Problems Using the Gibbs Sampler, Real Estate Economics 6 (3), Martin, R. J. (1984): Exact maximum likelihood for incomplete data from a correlated gaussian process, Communications in Statistics: Theory and Methods, 13, Tsutsumi, M. Yoshida, Y., Seya, H., Kawaguchi, Y.: Bayesian Spatial Modeling for Apartment Rent in Tokyo, the International Symposium on Statistical Analysis of Spatio-Temporal Data, pp.66-69, 006 (presented at the International Symposium on Statistical Analysis of Spatio-Temporal Data, The University of Tokyo, Tokyo, November 13-15, 006.) 13. Valente, J. Wu, S.S. Gelfand, A. and Sirmans, C. F.(005): Apartment rent prediction using spatial modeling, Journal of Real Estate Research, 7,

Application of eigenvector-based spatial filtering approach to. a multinomial logit model for land use data

Application of eigenvector-based spatial filtering approach to. a multinomial logit model for land use data Presented at the Seventh World Conference of the Spatial Econometrics Association, the Key Bridge Marriott Hotel, Washington, D.C., USA, July 10 12, 2013. Application of eigenvector-based spatial filtering

More information

Bayesian data analysis in practice: Three simple examples

Bayesian data analysis in practice: Three simple examples Bayesian data analysis in practice: Three simple examples Martin P. Tingley Introduction These notes cover three examples I presented at Climatea on 5 October 0. Matlab code is available by request to

More information

Cross-sectional space-time modeling using ARNN(p, n) processes

Cross-sectional space-time modeling using ARNN(p, n) processes Cross-sectional space-time modeling using ARNN(p, n) processes W. Polasek K. Kakamu September, 006 Abstract We suggest a new class of cross-sectional space-time models based on local AR models and nearest

More information

Influence of parameter estimation uncertainty in Kriging: Part 2 Test and case study applications

Influence of parameter estimation uncertainty in Kriging: Part 2 Test and case study applications Hydrology and Earth System Influence Sciences, of 5(), parameter 5 3 estimation (1) uncertainty EGS in Kriging: Part Test and case study applications Influence of parameter estimation uncertainty in Kriging:

More information

Measuring The Benefits of Air Quality Improvement: A Spatial Hedonic Approach. Chong Won Kim, Tim Phipps, and Luc Anselin

Measuring The Benefits of Air Quality Improvement: A Spatial Hedonic Approach. Chong Won Kim, Tim Phipps, and Luc Anselin Measuring The Benefits of Air Quality Improvement: A Spatial Hedonic Approach Chong Won Kim, Tim Phipps, and Luc Anselin Paper prepared for presentation at the AAEA annual meetings, Salt Lake City, August,

More information

Spatial Statistics For Real Estate Data 1

Spatial Statistics For Real Estate Data 1 1 Key words: spatial heterogeneity, spatial autocorrelation, spatial statistics, geostatistics, Geographical Information System SUMMARY: The paper presents spatial statistics tools in application to real

More information

Models for spatial data (cont d) Types of spatial data. Types of spatial data (cont d) Hierarchical models for spatial data

Models for spatial data (cont d) Types of spatial data. Types of spatial data (cont d) Hierarchical models for spatial data Hierarchical models for spatial data Based on the book by Banerjee, Carlin and Gelfand Hierarchical Modeling and Analysis for Spatial Data, 2004. We focus on Chapters 1, 2 and 5. Geo-referenced data arise

More information

Modeling Real Estate Data using Quantile Regression

Modeling Real Estate Data using Quantile Regression Modeling Real Estate Data using Semiparametric Quantile Regression Department of Statistics University of Innsbruck September 9th, 2011 Overview 1 Application: 2 3 4 Hedonic regression data for house prices

More information

ESTIMATING THE MEAN LEVEL OF FINE PARTICULATE MATTER: AN APPLICATION OF SPATIAL STATISTICS

ESTIMATING THE MEAN LEVEL OF FINE PARTICULATE MATTER: AN APPLICATION OF SPATIAL STATISTICS ESTIMATING THE MEAN LEVEL OF FINE PARTICULATE MATTER: AN APPLICATION OF SPATIAL STATISTICS Richard L. Smith Department of Statistics and Operations Research University of North Carolina Chapel Hill, N.C.,

More information

SUPPLEMENT TO MARKET ENTRY COSTS, PRODUCER HETEROGENEITY, AND EXPORT DYNAMICS (Econometrica, Vol. 75, No. 3, May 2007, )

SUPPLEMENT TO MARKET ENTRY COSTS, PRODUCER HETEROGENEITY, AND EXPORT DYNAMICS (Econometrica, Vol. 75, No. 3, May 2007, ) Econometrica Supplementary Material SUPPLEMENT TO MARKET ENTRY COSTS, PRODUCER HETEROGENEITY, AND EXPORT DYNAMICS (Econometrica, Vol. 75, No. 3, May 2007, 653 710) BY SANGHAMITRA DAS, MARK ROBERTS, AND

More information

Spatio-temporal precipitation modeling based on time-varying regressions

Spatio-temporal precipitation modeling based on time-varying regressions Spatio-temporal precipitation modeling based on time-varying regressions Oleg Makhnin Department of Mathematics New Mexico Tech Socorro, NM 87801 January 19, 2007 1 Abstract: A time-varying regression

More information

A SPATIAL ANALYSIS OF A RURAL LAND MARKET USING ALTERNATIVE SPATIAL WEIGHT MATRICES

A SPATIAL ANALYSIS OF A RURAL LAND MARKET USING ALTERNATIVE SPATIAL WEIGHT MATRICES A Spatial Analysis of a Rural Land Market Using Alternative Spatial Weight Matrices A SPATIAL ANALYSIS OF A RURAL LAND MARKET USING ALTERNATIVE SPATIAL WEIGHT MATRICES Patricia Soto, Louisiana State University

More information

Bayesian Inference: Probit and Linear Probability Models

Bayesian Inference: Probit and Linear Probability Models Utah State University DigitalCommons@USU All Graduate Plan B and other Reports Graduate Studies 5-1-2014 Bayesian Inference: Probit and Linear Probability Models Nate Rex Reasch Utah State University Follow

More information

Hedonic Housing Prices in Corsica: A hierarchical spatiotemporal approach WORKSHOP: THEORY AND PRACTICE OF SPDE MODELS AND INLA

Hedonic Housing Prices in Corsica: A hierarchical spatiotemporal approach WORKSHOP: THEORY AND PRACTICE OF SPDE MODELS AND INLA Hedonic Housing Prices in Corsica: A hierarchical spatiotemporal approach WORKSHOP: THEORY AND PRACTICE OF SPDE MODELS AND INLA LING Yuheng 1 30 Oct. 2018 1 PhD student in Economics - University of Corsica

More information

An Introduction to Pattern Statistics

An Introduction to Pattern Statistics An Introduction to Pattern Statistics Nearest Neighbors The CSR hypothesis Clark/Evans and modification Cuzick and Edwards and controls All events k function Weighted k function Comparative k functions

More information

I don t have much to say here: data are often sampled this way but we more typically model them in continuous space, or on a graph

I don t have much to say here: data are often sampled this way but we more typically model them in continuous space, or on a graph Spatial analysis Huge topic! Key references Diggle (point patterns); Cressie (everything); Diggle and Ribeiro (geostatistics); Dormann et al (GLMMs for species presence/abundance); Haining; (Pinheiro and

More information

INSTITUTE OF POLICY AND PLANNING SCIENCES. Discussion Paper Series

INSTITUTE OF POLICY AND PLANNING SCIENCES. Discussion Paper Series INSTITUTE OF POLICY AND PLANNING SCIENCES Discussion Paper Series No. 1102 Modeling with GIS: OD Commuting Times by Car and Public Transit in Tokyo by Mizuki Kawabata, Akiko Takahashi December, 2004 UNIVERSITY

More information

Method for Noise Addition for Individual Record Preserving Privacy and Statistical Characteristics: Case Study of Real Estate Transaction Data

Method for Noise Addition for Individual Record Preserving Privacy and Statistical Characteristics: Case Study of Real Estate Transaction Data CSIS Discussion Paper No. 140 Method for Noise Addition for Individual Record Preserving Privacy and Statistical Characteristics: Case Study of Real Estate Transaction Data Yuzo Maruyama Ryoko Tone and

More information

Departamento de Economía Universidad de Chile

Departamento de Economía Universidad de Chile Departamento de Economía Universidad de Chile GRADUATE COURSE SPATIAL ECONOMETRICS November 14, 16, 17, 20 and 21, 2017 Prof. Henk Folmer University of Groningen Objectives The main objective of the course

More information

Bayesian Methods for Machine Learning

Bayesian Methods for Machine Learning Bayesian Methods for Machine Learning CS 584: Big Data Analytics Material adapted from Radford Neal s tutorial (http://ftp.cs.utoronto.ca/pub/radford/bayes-tut.pdf), Zoubin Ghahramni (http://hunch.net/~coms-4771/zoubin_ghahramani_bayesian_learning.pdf),

More information

PhD/MA Econometrics Examination. January, 2015 PART A. (Answer any TWO from Part A)

PhD/MA Econometrics Examination. January, 2015 PART A. (Answer any TWO from Part A) PhD/MA Econometrics Examination January, 2015 Total Time: 8 hours MA students are required to answer from A and B. PhD students are required to answer from A, B, and C. PART A (Answer any TWO from Part

More information

PREDICTING HOUSE PRICES FOR SINGAPORE CONDOMINIUM RESALE MARKET: A COMPARISON OF TWO MODELS

PREDICTING HOUSE PRICES FOR SINGAPORE CONDOMINIUM RESALE MARKET: A COMPARISON OF TWO MODELS PREDICTING HOUSE PRICES FOR SINGAPORE CONDOMINIUM RESALE MARKET: A COMPARISON OF TWO MODELS ZHOU QIN (B. S. Yunnan University) A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF ESTATE MANAGEMENT DEPARTMENT

More information

Session 5B: A worked example EGARCH model

Session 5B: A worked example EGARCH model Session 5B: A worked example EGARCH model John Geweke Bayesian Econometrics and its Applications August 7, worked example EGARCH model August 7, / 6 EGARCH Exponential generalized autoregressive conditional

More information

Spatial Relationships in Rural Land Markets with Emphasis on a Flexible. Weights Matrix

Spatial Relationships in Rural Land Markets with Emphasis on a Flexible. Weights Matrix Spatial Relationships in Rural Land Markets with Emphasis on a Flexible Weights Matrix Patricia Soto, Lonnie Vandeveer, and Steve Henning Department of Agricultural Economics and Agribusiness Louisiana

More information

On dealing with spatially correlated residuals in remote sensing and GIS

On dealing with spatially correlated residuals in remote sensing and GIS On dealing with spatially correlated residuals in remote sensing and GIS Nicholas A. S. Hamm 1, Peter M. Atkinson and Edward J. Milton 3 School of Geography University of Southampton Southampton SO17 3AT

More information

Spatial Data Mining. Regression and Classification Techniques

Spatial Data Mining. Regression and Classification Techniques Spatial Data Mining Regression and Classification Techniques 1 Spatial Regression and Classisfication Discrete class labels (left) vs. continues quantities (right) measured at locations (2D for geographic

More information

Urban White Paper on Tokyo Metropolis 2002

Urban White Paper on Tokyo Metropolis 2002 Urban White Paper on Tokyo Metropolis 2002 By Bureau of City Planning Tokyo Metropolitan Government Part I. "Progress in IT and City Building" Effects of computer networks on cities and cities' response

More information

Lecture 4: Heteroskedasticity

Lecture 4: Heteroskedasticity Lecture 4: Heteroskedasticity Econometric Methods Warsaw School of Economics (4) Heteroskedasticity 1 / 24 Outline 1 What is heteroskedasticity? 2 Testing for heteroskedasticity White Goldfeld-Quandt Breusch-Pagan

More information

Kazuhiko Kakamu Department of Economics Finance, Institute for Advanced Studies. Abstract

Kazuhiko Kakamu Department of Economics Finance, Institute for Advanced Studies. Abstract Bayesian Estimation of A Distance Functional Weight Matrix Model Kazuhiko Kakamu Department of Economics Finance, Institute for Advanced Studies Abstract This paper considers the distance functional weight

More information

A Study on the Hedonic Analysis of Housing Market in the U-City

A Study on the Hedonic Analysis of Housing Market in the U-City A Study on the Hedonic Analysis of Housing Market in the Jiseon. Kim, Kabsung. Kim, and Seungbee. Choi Abstract The concept of is to integrate ubiquitous information technology services to our city space

More information

Spatial statistics, addition to Part I. Parameter estimation and kriging for Gaussian random fields

Spatial statistics, addition to Part I. Parameter estimation and kriging for Gaussian random fields Spatial statistics, addition to Part I. Parameter estimation and kriging for Gaussian random fields 1 Introduction Jo Eidsvik Department of Mathematical Sciences, NTNU, Norway. (joeid@math.ntnu.no) February

More information

House Price Distribution and Price Indexes in Tokyo

House Price Distribution and Price Indexes in Tokyo House Price Distribution and Price Indexes in Tokyo Singapore October 16, 2015 Yongheng Deng (IRES, NUS) with Xiangyu Guo (IRES, NUS) Daniel McMillan (Illinois University) Chihiro Shimizu (IRES, NUS) Institute

More information

Combining Regressive and Auto-Regressive Models for Spatial-Temporal Prediction

Combining Regressive and Auto-Regressive Models for Spatial-Temporal Prediction Combining Regressive and Auto-Regressive Models for Spatial-Temporal Prediction Dragoljub Pokrajac DPOKRAJA@EECS.WSU.EDU Zoran Obradovic ZORAN@EECS.WSU.EDU School of Electrical Engineering and Computer

More information

Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes

Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Sudipto Banerjee 1 and Andrew O. Finley 2 1 Biostatistics, School of Public Health, University of Minnesota,

More information

An Introduction to Spatial Statistics. Chunfeng Huang Department of Statistics, Indiana University

An Introduction to Spatial Statistics. Chunfeng Huang Department of Statistics, Indiana University An Introduction to Spatial Statistics Chunfeng Huang Department of Statistics, Indiana University Microwave Sounding Unit (MSU) Anomalies (Monthly): 1979-2006. Iron Ore (Cressie, 1986) Raw percent data

More information

Gaussian Process Regression Model in Spatial Logistic Regression

Gaussian Process Regression Model in Spatial Logistic Regression Journal of Physics: Conference Series PAPER OPEN ACCESS Gaussian Process Regression Model in Spatial Logistic Regression To cite this article: A Sofro and A Oktaviarina 018 J. Phys.: Conf. Ser. 947 01005

More information

BAYESIAN MODEL FOR SPATIAL DEPENDANCE AND PREDICTION OF TUBERCULOSIS

BAYESIAN MODEL FOR SPATIAL DEPENDANCE AND PREDICTION OF TUBERCULOSIS BAYESIAN MODEL FOR SPATIAL DEPENDANCE AND PREDICTION OF TUBERCULOSIS Srinivasan R and Venkatesan P Dept. of Statistics, National Institute for Research Tuberculosis, (Indian Council of Medical Research),

More information

Evaluating the Impact of the Fukushima Daiichi Nuclear Power Plant Accident

Evaluating the Impact of the Fukushima Daiichi Nuclear Power Plant Accident Evaluating the Impact of the Fukushima Daiichi Nuclear Power Plant Accident 1 Hirotaka Kato Graduate School of Economics, Kyoto University Yoshifumi Sako Graduate School of Economics, University of Tokyo

More information

Hastings-within-Gibbs Algorithm: Introduction and Application on Hierarchical Model

Hastings-within-Gibbs Algorithm: Introduction and Application on Hierarchical Model UNIVERSITY OF TEXAS AT SAN ANTONIO Hastings-within-Gibbs Algorithm: Introduction and Application on Hierarchical Model Liang Jing April 2010 1 1 ABSTRACT In this paper, common MCMC algorithms are introduced

More information

Monte Carlo Simulation. CWR 6536 Stochastic Subsurface Hydrology

Monte Carlo Simulation. CWR 6536 Stochastic Subsurface Hydrology Monte Carlo Simulation CWR 6536 Stochastic Subsurface Hydrology Steps in Monte Carlo Simulation Create input sample space with known distribution, e.g. ensemble of all possible combinations of v, D, q,

More information

Evaluating sustainable transportation offers through housing price: a comparative analysis of Nantes urban and periurban/rural areas (France)

Evaluating sustainable transportation offers through housing price: a comparative analysis of Nantes urban and periurban/rural areas (France) Evaluating sustainable transportation offers through housing price: a comparative analysis of Nantes urban and periurban/rural areas (France) Julie Bulteau, UVSQ-CEARC-OVSQ Thierry Feuillet, Université

More information

Markov Chain Monte Carlo methods

Markov Chain Monte Carlo methods Markov Chain Monte Carlo methods By Oleg Makhnin 1 Introduction a b c M = d e f g h i 0 f(x)dx 1.1 Motivation 1.1.1 Just here Supresses numbering 1.1.2 After this 1.2 Literature 2 Method 2.1 New math As

More information

Geographically weighted regression approach for origin-destination flows

Geographically weighted regression approach for origin-destination flows Geographically weighted regression approach for origin-destination flows Kazuki Tamesue 1 and Morito Tsutsumi 2 1 Graduate School of Information and Engineering, University of Tsukuba 1-1-1 Tennodai, Tsukuba,

More information

Statistícal Methods for Spatial Data Analysis

Statistícal Methods for Spatial Data Analysis Texts in Statistícal Science Statistícal Methods for Spatial Data Analysis V- Oliver Schabenberger Carol A. Gotway PCT CHAPMAN & K Contents Preface xv 1 Introduction 1 1.1 The Need for Spatial Analysis

More information

Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes

Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Alan Gelfand 1 and Andrew O. Finley 2 1 Department of Statistical Science, Duke University, Durham, North

More information

Chapter 4 - Fundamentals of spatial processes Lecture notes

Chapter 4 - Fundamentals of spatial processes Lecture notes Chapter 4 - Fundamentals of spatial processes Lecture notes Geir Storvik January 21, 2013 STK4150 - Intro 2 Spatial processes Typically correlation between nearby sites Mostly positive correlation Negative

More information

How does Transport Infrastructure Affect Dwelling Prices in Athens?

How does Transport Infrastructure Affect Dwelling Prices in Athens? How does Transport Infrastructure Affect Dwelling Prices in Athens? Dimitrios Efthymiou* School of Rural and Surveying Engineering, National Technical University of Athens, Greece Constantinos Antoniou

More information

Bayesian Dynamic Modeling for Space-time Data in R

Bayesian Dynamic Modeling for Space-time Data in R Bayesian Dynamic Modeling for Space-time Data in R Andrew O. Finley and Sudipto Banerjee September 5, 2014 We make use of several libraries in the following example session, including: ˆ library(fields)

More information

Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes

Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Andrew O. Finley Department of Forestry & Department of Geography, Michigan State University, Lansing

More information

Bayesian spatial hierarchical modeling for temperature extremes

Bayesian spatial hierarchical modeling for temperature extremes Bayesian spatial hierarchical modeling for temperature extremes Indriati Bisono Dr. Andrew Robinson Dr. Aloke Phatak Mathematics and Statistics Department The University of Melbourne Maths, Informatics

More information

An Econometric Analysis of Determinants of House Rents in Istanbul

An Econometric Analysis of Determinants of House Rents in Istanbul An Econometric Analysis of Determinants of House Rents in Istanbul Ebru Çağlayan- Akay 1* Engin Bekar 1 M.Hanifi Van 2 1.Marmara University, Department of Econometrics, Istanbul, Turkey 2.Yüzüncü Yıl University,

More information

Statistics for analyzing and modeling precipitation isotope ratios in IsoMAP

Statistics for analyzing and modeling precipitation isotope ratios in IsoMAP Statistics for analyzing and modeling precipitation isotope ratios in IsoMAP The IsoMAP uses the multiple linear regression and geostatistical methods to analyze isotope data Suppose the response variable

More information

ECON 4230 Intermediate Econometric Theory Exam

ECON 4230 Intermediate Econometric Theory Exam ECON 4230 Intermediate Econometric Theory Exam Multiple Choice (20 pts). Circle the best answer. 1. The Classical assumption of mean zero errors is satisfied if the regression model a) is linear in the

More information

Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes

Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Andrew O. Finley 1 and Sudipto Banerjee 2 1 Department of Forestry & Department of Geography, Michigan

More information

Lecture 5: Spatial probit models. James P. LeSage University of Toledo Department of Economics Toledo, OH

Lecture 5: Spatial probit models. James P. LeSage University of Toledo Department of Economics Toledo, OH Lecture 5: Spatial probit models James P. LeSage University of Toledo Department of Economics Toledo, OH 43606 jlesage@spatial-econometrics.com March 2004 1 A Bayesian spatial probit model with individual

More information

Ref.: Spring SOS3003 Applied data analysis for social science Lecture note

Ref.:   Spring SOS3003 Applied data analysis for social science Lecture note SOS3003 Applied data analysis for social science Lecture note 05-2010 Erling Berge Department of sociology and political science NTNU Spring 2010 Erling Berge 2010 1 Literature Regression criticism I Hamilton

More information

Christopher Dougherty London School of Economics and Political Science

Christopher Dougherty London School of Economics and Political Science Introduction to Econometrics FIFTH EDITION Christopher Dougherty London School of Economics and Political Science OXFORD UNIVERSITY PRESS Contents INTRODU CTION 1 Why study econometrics? 1 Aim of this

More information

Econometrics Review questions for exam

Econometrics Review questions for exam Econometrics Review questions for exam Nathaniel Higgins nhiggins@jhu.edu, 1. Suppose you have a model: y = β 0 x 1 + u You propose the model above and then estimate the model using OLS to obtain: ŷ =

More information

Apartment Rent Prediction Using Spatial Modeling

Apartment Rent Prediction Using Spatial Modeling Apartment Rent Prediction Using Spatial Modeling Authors James Valente, ShanShan Wu, Alan Gelfand and C. F. Sirmans Abstract This paper provides a new model to explain local variation in apartment rents

More information

Econometrics. 9) Heteroscedasticity and autocorrelation

Econometrics. 9) Heteroscedasticity and autocorrelation 30C00200 Econometrics 9) Heteroscedasticity and autocorrelation Timo Kuosmanen Professor, Ph.D. http://nomepre.net/index.php/timokuosmanen Today s topics Heteroscedasticity Possible causes Testing for

More information

Predicting House Prices with Spatial Dependence: A Comparison of Alternative Methods

Predicting House Prices with Spatial Dependence: A Comparison of Alternative Methods Predicting House Prices with Spatial Dependence: A Comparison of Alternative Methods Authors Steven C. Bourassa, Eva Cantoni, and Martin Hoesli Abstract This paper compares alternative methods for taking

More information

Department of Economics, UCSB UC Santa Barbara

Department of Economics, UCSB UC Santa Barbara Department of Economics, UCSB UC Santa Barbara Title: Past trend versus future expectation: test of exchange rate volatility Author: Sengupta, Jati K., University of California, Santa Barbara Sfeir, Raymond,

More information

CBMS Lecture 1. Alan E. Gelfand Duke University

CBMS Lecture 1. Alan E. Gelfand Duke University CBMS Lecture 1 Alan E. Gelfand Duke University Introduction to spatial data and models Researchers in diverse areas such as climatology, ecology, environmental exposure, public health, and real estate

More information

(5) Multi-parameter models - Gibbs sampling. ST440/540: Applied Bayesian Analysis

(5) Multi-parameter models - Gibbs sampling. ST440/540: Applied Bayesian Analysis Summarizing a posterior Given the data and prior the posterior is determined Summarizing the posterior gives parameter estimates, intervals, and hypothesis tests Most of these computations are integrals

More information

Marginal Specifications and a Gaussian Copula Estimation

Marginal Specifications and a Gaussian Copula Estimation Marginal Specifications and a Gaussian Copula Estimation Kazim Azam Abstract Multivariate analysis involving random variables of different type like count, continuous or mixture of both is frequently required

More information

Introduction. Introduction (Contd.) Market Equilibrium and Spatial Variability in the Value of Housing Attributes. Urban location theory.

Introduction. Introduction (Contd.) Market Equilibrium and Spatial Variability in the Value of Housing Attributes. Urban location theory. Forestry, Wildlife, and Fisheries Graduate Seminar Market Equilibrium and Spatial Variability in the Value of Housing Attributes Seung Gyu Kim Wednesday, 12 March 2008 1 Introduction Urban location theory

More information

2. Linear regression with multiple regressors

2. Linear regression with multiple regressors 2. Linear regression with multiple regressors Aim of this section: Introduction of the multiple regression model OLS estimation in multiple regression Measures-of-fit in multiple regression Assumptions

More information

Introduction to Econometrics

Introduction to Econometrics Introduction to Econometrics T H I R D E D I T I O N Global Edition James H. Stock Harvard University Mark W. Watson Princeton University Boston Columbus Indianapolis New York San Francisco Upper Saddle

More information

Multilevel modeling and panel data analysis in educational research (Case study: National examination data senior high school in West Java)

Multilevel modeling and panel data analysis in educational research (Case study: National examination data senior high school in West Java) Multilevel modeling and panel data analysis in educational research (Case study: National examination data senior high school in West Java) Pepi Zulvia, Anang Kurnia, and Agus M. Soleh Citation: AIP Conference

More information

Introduction to Spatial Data and Models

Introduction to Spatial Data and Models Introduction to Spatial Data and Models Sudipto Banerjee 1 and Andrew O. Finley 2 1 Biostatistics, School of Public Health, University of Minnesota, Minneapolis, Minnesota, U.S.A. 2 Department of Forestry

More information

Online appendix to On the stability of the excess sensitivity of aggregate consumption growth in the US

Online appendix to On the stability of the excess sensitivity of aggregate consumption growth in the US Online appendix to On the stability of the excess sensitivity of aggregate consumption growth in the US Gerdie Everaert 1, Lorenzo Pozzi 2, and Ruben Schoonackers 3 1 Ghent University & SHERPPA 2 Erasmus

More information

Predicting House Prices with Spatial Dependence: A Comparison of Alternative Methods

Predicting House Prices with Spatial Dependence: A Comparison of Alternative Methods Predicting House Prices with Spatial Dependence: A Comparison of Alternative Methods Steven C. Bourassa School of Urban and Public Affairs, University of Louisville, 426 W. Bloom Street, Louisville, Kentucky

More information

iafor The International Academic Forum

iafor The International Academic Forum Land Use Change with Externalities in the Fringe of Jakarta Metropolitan: Spatial Tobit Model Rahma Fitriani, University of Brawijaya, Indonesia Eni Sumarminingsih, University of Brawijaya, Indonesia The

More information

Report. Reference. Spatial Dependence, Housing Submarkets and House Price Prediction. BOURASSA, Steven C., CANTONI, Eva, HOESLI, Martin E.

Report. Reference. Spatial Dependence, Housing Submarkets and House Price Prediction. BOURASSA, Steven C., CANTONI, Eva, HOESLI, Martin E. Report Spatial Dependence, Housing Submarkets and House Price Prediction BOURASSA, Steven C., CANTONI, Eva, HOESLI, Martin E. Abstract This paper compares alternative methods of controlling for the spatial

More information

Least Squares Estimation-Finite-Sample Properties

Least Squares Estimation-Finite-Sample Properties Least Squares Estimation-Finite-Sample Properties Ping Yu School of Economics and Finance The University of Hong Kong Ping Yu (HKU) Finite-Sample 1 / 29 Terminology and Assumptions 1 Terminology and Assumptions

More information

Introduction to Bayesian Statistics and Markov Chain Monte Carlo Estimation. EPSY 905: Multivariate Analysis Spring 2016 Lecture #10: April 6, 2016

Introduction to Bayesian Statistics and Markov Chain Monte Carlo Estimation. EPSY 905: Multivariate Analysis Spring 2016 Lecture #10: April 6, 2016 Introduction to Bayesian Statistics and Markov Chain Monte Carlo Estimation EPSY 905: Multivariate Analysis Spring 2016 Lecture #10: April 6, 2016 EPSY 905: Intro to Bayesian and MCMC Today s Class An

More information

Making sense of Econometrics: Basics

Making sense of Econometrics: Basics Making sense of Econometrics: Basics Lecture 7: Multicollinearity Egypt Scholars Economic Society November 22, 2014 Assignment & feedback Multicollinearity enter classroom at room name c28efb78 http://b.socrative.com/login/student/

More information

A Test of Cointegration Rank Based Title Component Analysis.

A Test of Cointegration Rank Based Title Component Analysis. A Test of Cointegration Rank Based Title Component Analysis Author(s) Chigira, Hiroaki Citation Issue 2006-01 Date Type Technical Report Text Version publisher URL http://hdl.handle.net/10086/13683 Right

More information

SPACE Workshop NSF NCGIA CSISS UCGIS SDSU. Aldstadt, Getis, Jankowski, Rey, Weeks SDSU F. Goodchild, M. Goodchild, Janelle, Rebich UCSB

SPACE Workshop NSF NCGIA CSISS UCGIS SDSU. Aldstadt, Getis, Jankowski, Rey, Weeks SDSU F. Goodchild, M. Goodchild, Janelle, Rebich UCSB SPACE Workshop NSF NCGIA CSISS UCGIS SDSU Aldstadt, Getis, Jankowski, Rey, Weeks SDSU F. Goodchild, M. Goodchild, Janelle, Rebich UCSB August 2-8, 2004 San Diego State University Some Examples of Spatial

More information

A STATISTICAL TECHNIQUE FOR MODELLING NON-STATIONARY SPATIAL PROCESSES

A STATISTICAL TECHNIQUE FOR MODELLING NON-STATIONARY SPATIAL PROCESSES A STATISTICAL TECHNIQUE FOR MODELLING NON-STATIONARY SPATIAL PROCESSES JOHN STEPHENSON 1, CHRIS HOLMES, KERRY GALLAGHER 1 and ALEXANDRE PINTORE 1 Dept. Earth Science and Engineering, Imperial College,

More information

MCMC algorithms for fitting Bayesian models

MCMC algorithms for fitting Bayesian models MCMC algorithms for fitting Bayesian models p. 1/1 MCMC algorithms for fitting Bayesian models Sudipto Banerjee sudiptob@biostat.umn.edu University of Minnesota MCMC algorithms for fitting Bayesian models

More information

Trend Report of the Values of Intensively Used Land in Major Cities - Land Value LOOK Report -

Trend Report of the Values of Intensively Used Land in Major Cities - Land Value LOOK Report - Trend Report of the Values of Intensively Used Land in Major Cities - Land Value LOOK Report - 44 th Issue - Third Quarter of 2018 Trend from July 1, 2018 to October 1, 2018 Land Price Research Division

More information

LECTURE 10. Introduction to Econometrics. Multicollinearity & Heteroskedasticity

LECTURE 10. Introduction to Econometrics. Multicollinearity & Heteroskedasticity LECTURE 10 Introduction to Econometrics Multicollinearity & Heteroskedasticity November 22, 2016 1 / 23 ON PREVIOUS LECTURES We discussed the specification of a regression equation Specification consists

More information

The Built Environment, Car Ownership, and Travel Behavior in Seoul

The Built Environment, Car Ownership, and Travel Behavior in Seoul The Built Environment, Car Ownership, and Travel Behavior in Seoul Sang-Kyu Cho, Ph D. Candidate So-Ra Baek, Master Course Student Seoul National University Abstract Although the idea of integrating land

More information

Areal data models. Spatial smoothers. Brook s Lemma and Gibbs distribution. CAR models Gaussian case Non-Gaussian case

Areal data models. Spatial smoothers. Brook s Lemma and Gibbs distribution. CAR models Gaussian case Non-Gaussian case Areal data models Spatial smoothers Brook s Lemma and Gibbs distribution CAR models Gaussian case Non-Gaussian case SAR models Gaussian case Non-Gaussian case CAR vs. SAR STAR models Inference for areal

More information

27 th Issue - Second Quarter of 2014

27 th Issue - Second Quarter of 2014 27 th Issue - Second Quarter of 2014 Trend from April 1, 2014 to July 1, 2014 Land Price Research Division Ministry of Land, Infrastructure, Transport and Tourism August 2014 Survey Outline 1. Survey

More information

Warwick Business School Forecasting System. Summary. Ana Galvao, Anthony Garratt and James Mitchell November, 2014

Warwick Business School Forecasting System. Summary. Ana Galvao, Anthony Garratt and James Mitchell November, 2014 Warwick Business School Forecasting System Summary Ana Galvao, Anthony Garratt and James Mitchell November, 21 The main objective of the Warwick Business School Forecasting System is to provide competitive

More information

Regression Analysis. BUS 735: Business Decision Making and Research. Learn how to detect relationships between ordinal and categorical variables.

Regression Analysis. BUS 735: Business Decision Making and Research. Learn how to detect relationships between ordinal and categorical variables. Regression Analysis BUS 735: Business Decision Making and Research 1 Goals of this section Specific goals Learn how to detect relationships between ordinal and categorical variables. Learn how to estimate

More information

Introduction to Geostatistics

Introduction to Geostatistics Introduction to Geostatistics Abhi Datta 1, Sudipto Banerjee 2 and Andrew O. Finley 3 July 31, 2017 1 Department of Biostatistics, Bloomberg School of Public Health, Johns Hopkins University, Baltimore,

More information

Principles of Bayesian Inference

Principles of Bayesian Inference Principles of Bayesian Inference Sudipto Banerjee University of Minnesota July 20th, 2008 1 Bayesian Principles Classical statistics: model parameters are fixed and unknown. A Bayesian thinks of parameters

More information

Introduction to Regression Analysis. Dr. Devlina Chatterjee 11 th August, 2017

Introduction to Regression Analysis. Dr. Devlina Chatterjee 11 th August, 2017 Introduction to Regression Analysis Dr. Devlina Chatterjee 11 th August, 2017 What is regression analysis? Regression analysis is a statistical technique for studying linear relationships. One dependent

More information

Heteroskedasticity. y i = β 0 + β 1 x 1i + β 2 x 2i β k x ki + e i. where E(e i. ) σ 2, non-constant variance.

Heteroskedasticity. y i = β 0 + β 1 x 1i + β 2 x 2i β k x ki + e i. where E(e i. ) σ 2, non-constant variance. Heteroskedasticity y i = β + β x i + β x i +... + β k x ki + e i where E(e i ) σ, non-constant variance. Common problem with samples over individuals. ê i e ˆi x k x k AREC-ECON 535 Lec F Suppose y i =

More information

Spatiotemporal Analysis of Solar Radiation for Sustainable Research in the Presence of Uncertain Measurements

Spatiotemporal Analysis of Solar Radiation for Sustainable Research in the Presence of Uncertain Measurements Spatiotemporal Analysis of Solar Radiation for Sustainable Research in the Presence of Uncertain Measurements Alexander Kolovos SAS Institute, Inc. alexander.kolovos@sas.com Abstract. The study of incoming

More information

Measuring The Benefits of Urban Green Areas: A Spatial Hedonic Approach

Measuring The Benefits of Urban Green Areas: A Spatial Hedonic Approach Measuring The Benefits of Urban Green Areas: A Spatial Hedonic Approach An-Ming Wang Ph.D. Student, The Graduate Institute of Urban Planning, National Taipei University, Taipei, Taiwan Tel: 886-2-25009749

More information

INTRODUCTORY REGRESSION ANALYSIS

INTRODUCTORY REGRESSION ANALYSIS ;»»>? INTRODUCTORY REGRESSION ANALYSIS With Computer Application for Business and Economics Allen Webster Routledge Taylor & Francis Croup NEW YORK AND LONDON TABLE OF CONTENT IN DETAIL INTRODUCTORY REGRESSION

More information

VAR models with non-gaussian shocks

VAR models with non-gaussian shocks VAR models with non-gaussian shocks Ching-Wai (Jeremy) Chiu Haroon Mumtaz Gabor Pinter September 27, 2016 Motivation and aims Density of the residuals from a recursive VAR(13) (1960m1-2015m6) Motivation

More information

ECON3150/4150 Spring 2016

ECON3150/4150 Spring 2016 ECON3150/4150 Spring 2016 Lecture 6 Multiple regression model Siv-Elisabeth Skjelbred University of Oslo February 5th Last updated: February 3, 2016 1 / 49 Outline Multiple linear regression model and

More information

Predicting Spatial Patterns of House Prices Using LPR and Bayesian Smoothing

Predicting Spatial Patterns of House Prices Using LPR and Bayesian Smoothing 2002 V30 4: pp. 505 532 REAL ESTATE ECONOMICS Predicting Spatial Patterns of House Prices Using LPR and Bayesian Smoothing John M. Clapp, Hyon-Jung Kim and Alan E. Gelfand This article is motivated by

More information

Markov Chain Monte Carlo methods

Markov Chain Monte Carlo methods Markov Chain Monte Carlo methods Tomas McKelvey and Lennart Svensson Signal Processing Group Department of Signals and Systems Chalmers University of Technology, Sweden November 26, 2012 Today s learning

More information

Bayesian Semiparametric GARCH Models

Bayesian Semiparametric GARCH Models Bayesian Semiparametric GARCH Models Xibin (Bill) Zhang and Maxwell L. King Department of Econometrics and Business Statistics Faculty of Business and Economics xibin.zhang@monash.edu Quantitative Methods

More information