Motivated by a large multilevel survey data conducted by the US Veterans Health Adminis-tration (... more Motivated by a large multilevel survey data conducted by the US Veterans Health Adminis-tration (VHA), we propose a structural equations model which involves a set of latent variables to capture dependence between different responses, a set of facility level random effects to cap-ture ...
This paper estimates Spatial Bayesian Vector Autoregressive (SBVAR) models, based on the First-Or... more This paper estimates Spatial Bayesian Vector Autoregressive (SBVAR) models, based on the First-Order Spatial Contiguity and the Random Walk Averaging priors, for six metropolitan areas of South Africa, using monthly data over the period of 1993:07 to 2005:06. We then forecast one- to six-months-ahead house prices over the forecast horizon of 2005:07 to 2007:06. When we compare forecasts generated from the SBVARs with those from an unrestricted Vector Autoregressive (VAR) and the Bayesian Vector Autoregressive (BVAR) models based on the Minnesota prior, we find that the spatial models tend to outperform the other models for large middle-segment houses; while the VAR and the BVAR models tend to produce lower average out-of-sample forecast errors for middle and small-middle segment houses, respectively. In addition, based on the priors used to estimate the Bayesian models, our results also suggest that prices tend to converge for both large- and middle-sized houses, but no such evidence could be obtained for the small-sized houses.
... Page 3. Das,S.,Gupta,R.,Kaya,PA Convergence of Metropolitan House Prices in South Africa 175 ... more ... Page 3. Das,S.,Gupta,R.,Kaya,PA Convergence of Metropolitan House Prices in South Africa 175 ... However, Caner and Kilian (2000) have indicated that the KPSS test is found to have serious distortions. Given this, the choice of the efficient unit root tests proposed by Elliott et al. ...
Journal of Real Estate Finance and Economics, 2010
This paper estimates Bayesian Vector Autoregressive (BVAR) models, both spatial and non-spatial (... more This paper estimates Bayesian Vector Autoregressive (BVAR) models, both spatial and non-spatial (univariate and multivariate), for the twenty largest states of the US economy, using quarterly data over the period 1976:Q1–1994:Q4; and then forecasts one-to-four quarters-ahead real house price growth over the out-of-sample horizon of 1995:Q1–2006:Q4. The forecasts are evaluated by comparing them with those from an unrestricted classical Vector Autoregressive (VAR) model and the corresponding univariate variant of the same. Finally, the models that produce the minimum average Root Mean Square Errors (RMSEs), are used to predict the downturns in the real house price growth over the recent period of 2007:Q1–2008:Q1. The results show that the BVARs, in whatever form they might be, are the best performing models in 19 of the 20 states. Moreover, these models do a fair job in predicting the downturn in 18 of the 19 states.
1. Fire is an important process in Mediterranean-ecosystem shrublands, and prescribed burning is ... more 1. Fire is an important process in Mediterranean-ecosystem shrublands, and prescribed burning is often used to manage these ecosystems. Analyses of past fire regimes are required to interpret biotic responses to fire, as well as to assess the degree to which management interventions have been able to influence the fire regime.2. We used a spatial data base of fires within 10 protected areas covering >720 000 ha to examine the frequency, seasonality, size and cause of fires over four decades. Our study covered five fire climate zones and a range of mountain fynbos shrubland types. We examined whether regular prescribed burning would be necessary to rejuvenate the vegetation, and also to reduce the incidence and extent of wildfires.3. Cumulative fire frequency distributions indicated that the probability of fire was not strongly affected by post-fire age, with 50% of the area experiencing a successive fire within 10–13 years after the previous fire in most areas. This suggests that the accumulation of fuel did not limit the occurrence of wildfires, and that regular prescribed burning would not necessarily reduce the risk of wildfires.4. Inland zones experienced more severe fire weather than coastal zones (∼35% vs. 11–19% of days with high to very high fire danger, respectively). Despite these differences, fire return periods were similar (10–13 years), suggesting that the availability of ignitions, and not fuel or weather, limited the occurrence of wildfires.5. Despite a policy that promoted prescribed burning, a relatively small area (between 4·6% and 32·4% of the area of all fires) burned in prescribed burns. Seasonal restrictions for safety and ecological reasons, the imperative to integrate planned fires with invasive alien plant treatments and unplanned wildfires have all contributed to the relatively small area that burnt in prescribed burns.6. Synthesis and applications. Recurrent wildfires, and not prescribed burning, are providing sufficient opportunities for fire-stimulated regeneration in fynbos ecosystems. Because of this, and because burning to reduce fuel loads is unlikely to prevent wildfires, there should be less pressure to conduct prescribed burning. The predicted growth in human populations in all areas is expected to increase the number of ignition opportunities and the frequency of fires, with detrimental consequences for biodiversity conservation and the control of invasive alien trees. Fire frequency should thus be monitored and steps should be taken to protect areas that burn too frequently.
Motivated by a large multilevel survey data conducted by the US Veterans Health Adminis-tration (... more Motivated by a large multilevel survey data conducted by the US Veterans Health Adminis-tration (VHA), we propose a structural equations model which involves a set of latent variables to capture dependence between different responses, a set of facility level random effects to cap-ture ...
This paper estimates Spatial Bayesian Vector Autoregressive (SBVAR) models, based on the First-Or... more This paper estimates Spatial Bayesian Vector Autoregressive (SBVAR) models, based on the First-Order Spatial Contiguity and the Random Walk Averaging priors, for six metropolitan areas of South Africa, using monthly data over the period of 1993:07 to 2005:06. We then forecast one- to six-months-ahead house prices over the forecast horizon of 2005:07 to 2007:06. When we compare forecasts generated from the SBVARs with those from an unrestricted Vector Autoregressive (VAR) and the Bayesian Vector Autoregressive (BVAR) models based on the Minnesota prior, we find that the spatial models tend to outperform the other models for large middle-segment houses; while the VAR and the BVAR models tend to produce lower average out-of-sample forecast errors for middle and small-middle segment houses, respectively. In addition, based on the priors used to estimate the Bayesian models, our results also suggest that prices tend to converge for both large- and middle-sized houses, but no such evidence could be obtained for the small-sized houses.
... Page 3. Das,S.,Gupta,R.,Kaya,PA Convergence of Metropolitan House Prices in South Africa 175 ... more ... Page 3. Das,S.,Gupta,R.,Kaya,PA Convergence of Metropolitan House Prices in South Africa 175 ... However, Caner and Kilian (2000) have indicated that the KPSS test is found to have serious distortions. Given this, the choice of the efficient unit root tests proposed by Elliott et al. ...
Journal of Real Estate Finance and Economics, 2010
This paper estimates Bayesian Vector Autoregressive (BVAR) models, both spatial and non-spatial (... more This paper estimates Bayesian Vector Autoregressive (BVAR) models, both spatial and non-spatial (univariate and multivariate), for the twenty largest states of the US economy, using quarterly data over the period 1976:Q1–1994:Q4; and then forecasts one-to-four quarters-ahead real house price growth over the out-of-sample horizon of 1995:Q1–2006:Q4. The forecasts are evaluated by comparing them with those from an unrestricted classical Vector Autoregressive (VAR) model and the corresponding univariate variant of the same. Finally, the models that produce the minimum average Root Mean Square Errors (RMSEs), are used to predict the downturns in the real house price growth over the recent period of 2007:Q1–2008:Q1. The results show that the BVARs, in whatever form they might be, are the best performing models in 19 of the 20 states. Moreover, these models do a fair job in predicting the downturn in 18 of the 19 states.
1. Fire is an important process in Mediterranean-ecosystem shrublands, and prescribed burning is ... more 1. Fire is an important process in Mediterranean-ecosystem shrublands, and prescribed burning is often used to manage these ecosystems. Analyses of past fire regimes are required to interpret biotic responses to fire, as well as to assess the degree to which management interventions have been able to influence the fire regime.2. We used a spatial data base of fires within 10 protected areas covering >720 000 ha to examine the frequency, seasonality, size and cause of fires over four decades. Our study covered five fire climate zones and a range of mountain fynbos shrubland types. We examined whether regular prescribed burning would be necessary to rejuvenate the vegetation, and also to reduce the incidence and extent of wildfires.3. Cumulative fire frequency distributions indicated that the probability of fire was not strongly affected by post-fire age, with 50% of the area experiencing a successive fire within 10–13 years after the previous fire in most areas. This suggests that the accumulation of fuel did not limit the occurrence of wildfires, and that regular prescribed burning would not necessarily reduce the risk of wildfires.4. Inland zones experienced more severe fire weather than coastal zones (∼35% vs. 11–19% of days with high to very high fire danger, respectively). Despite these differences, fire return periods were similar (10–13 years), suggesting that the availability of ignitions, and not fuel or weather, limited the occurrence of wildfires.5. Despite a policy that promoted prescribed burning, a relatively small area (between 4·6% and 32·4% of the area of all fires) burned in prescribed burns. Seasonal restrictions for safety and ecological reasons, the imperative to integrate planned fires with invasive alien plant treatments and unplanned wildfires have all contributed to the relatively small area that burnt in prescribed burns.6. Synthesis and applications. Recurrent wildfires, and not prescribed burning, are providing sufficient opportunities for fire-stimulated regeneration in fynbos ecosystems. Because of this, and because burning to reduce fuel loads is unlikely to prevent wildfires, there should be less pressure to conduct prescribed burning. The predicted growth in human populations in all areas is expected to increase the number of ignition opportunities and the frequency of fires, with detrimental consequences for biodiversity conservation and the control of invasive alien trees. Fire frequency should thus be monitored and steps should be taken to protect areas that burn too frequently.
Uploads
Papers by Sonali Das