|
on Computational Economics |
|
Issue of 2014‒08‒02
seven papers chosen by |
| By: | Meijers H.H.M.; Nomaler Z.O.; Verspagen B. (UNU-MERIT) |
| Abstract: | We develop a microsimulation model for the macroeconomic business cycle. Our model is based on three main ideas i we want to specify how macroeconomic coordination is achieved without a dominating influence of price mechanisms, ii we want to incorporate the stock-flow-consistent approach that has become popular in post-Keynesian macroeconomics, and iii we want to allow for bankruptcies as a major mechanism in the business cycle. Compared to existing stock-flow-consistent models, our model has relatively few equations. It is operationalized using micro, agent-based simulation. The results show a clear business cycle that is driven by accumulation of financial assets and the effects this has on the real economy. By changing some of the key parameters, we show how the nature of the business cycle changes as a result of changes in the assumed behaviour of agents. |
| Keywords: | Current Heterodox Approaches: General; Current Heterodox Approaches: Institutional; Evolutionary; Macroeconomics and Monetary Economics: General; Business Fluctuations; Cycles; |
| JEL: | E00 E32 B50 B52 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:unm:unumer:2014047&r=cmp |
| By: | Roberto Casarin (University Ca’ Foscari of Venice and GRETA); Stefano Grassi (School of Economics University of Kent and CREATES); Francesco Ravazzolo (Norges Bank (Central Bank of Norway) and BI Norwegian Business School); Harman K. van Dijk (Erasmus University Rotterdam VU University Amsterdam and Tinbergen Institute) |
| Abstract: | This paper presents the MATLAB package DeCo (density combination) which is based on the paper by Billio, Casarin, Ravazzolo, and van Dijk (2013) where a constructive Bayesian approach is presented for combining predictive densities originating from different models or other sources of information. The combination weights are time-varying and may depend on past predictive forecasting performances and other learning mechanisms. The core algorithm is the function DeCo which applies banks of parallel Sequential Monte Carlo algorithms to filter the time-varying combination weights. The DeCo procedure has been implemented both for standard CPU computing and for graphical process unit (GPU) parallel computing. For the GPU implementation we use the MATLAB parallel computing toolbox and show how to use general purpose GPU computing almost effortlessly. This GPU implementation provides a speed up of the execution time of up to seventy times on a standard CPU MATLAB implementation on a multicore CPU. We show the use of the package and the computational gain of the GPU version through some simulation experiments and empirical applications. |
| Keywords: | Density forecast combination, sequrntial Monte Carlo, parallel computing, GPU, MATLAB |
| Date: | 2014–07–24 |
| URL: | http://d.repec.org/n?u=RePEc:bno:worpap:2014_11&r=cmp |
| By: | Maier, Gunther |
| Abstract: | The hedonic price method is well established as a method for deriving real estate price indices and for identifying the factors that influence real estate prices in the market. Despite a close relationship to the traditional sales comparison approach and conceptual and statistical advantages, the hedonic price method is used much less in the appraisal of individual objects. In recent years, locational facors have been increasingly included into the set of explanatory variables as well as into the specification of the error term. This has not only improved the explanatory power of the method, but also introduced additional challenges for the interpretation of results and particularly for the use of these results for the appraisal of individual objects.The paper discusses the hedonic price approach to real estate appraisal and especially its relation to the traditional sales comparison approach. It will be shown that spatial econometric methods can be applied to improve the power of the hedonic price regression. The main part of the paper will be devoted to the question of how to utilize the results of the hedonic regression for appraisal of individual objects, what challenges the different possible specifications of the regression equation on the one hand and of the structure of the error term on the other hand raise. The paper presents algorithms for handling these computational issues and presents simulation results for each case. |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:arz:wpaper:eres2014_54&r=cmp |
| By: | Kovacs, Alexandra; Wiegand, Dietmar; Emrich, Stefan |
| Abstract: | The development of corporate buildings and the structure of workplaces changed dramatically during the last number of decades. Studies show that the utilization of e.g. office space lies at 5% over its lifecycle – so we need to rethink the design & the management of corporate facilities. In fact buildings which are heated, cooled and maintained but remain largely unused over time are economically & ecologically inefficient! Our built environment is identified as a crucial area for the realisation of the EU climate goals but the potential to reduce energy-demand and CO2-Emissions remain widely unused. Beside the opportunity to save a lot of (real estate and company/organisational) money through an optimized FM - the mentioned arguments above illustrate, that the improvement of space-utilization is also part of Corporate Social Responsibility. More-Space analyses saving potentials of e.g. flexible workplace management. Know-How from real estate development, facility management and mathematical modelling and simulation are brought together to improve and optimize space-utilisation. On going studies show impressive saving potentials up to 50% of the current workplaces and an optimised utilisation up to 80% within companies. Other studies within school buildings showed that up to 40% of the classrooms could be used for other purposes through improved space-management. These evaluations add up to massive saving potentials of user costs of buildings and tons of CO2-Emissions per year. The More-Space Tool is able to demonstrate each company its maximum benefit, as the model is customizable and supports rapid prototyping – a relevant characteristic for decision makers as managing directors, facility managers and others. The More-Space approach underlines that a successful achievement of the FM-potentials needs interaction between at least four competences: 1.Organisational development – necessary changes of processes and organizational structures; 2.Space Management - potentials e.g. within intelligent distribution of fixed-term rights of utilization but also within spontaneous occupancy; 3.Real Estate Planning/Development – Developers and designers must be able to create built infrastructures, which support dynamic usage of buildings and 4.Simulation – the high potentials for more efficient utilization often remain undetected until dynamic event-based computer-simulations facilitate, as the complexity of dynamic viewing of utilisation is extremely high. |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:arz:wpaper:eres2013_123&r=cmp |
| By: | Rengarajan, Satyanarain; Hin, David Ho Kim |
| Abstract: | Throughout the OECD world and beyond, localised cluster based initiatives have increasingly being seen as the main industrial policy option to sustain regional competitiveness and economic prosperity (OECD, 2000). Industrialized nations in particular have come up with large scale plans to develop what is known in the literature as the 'Knowledge Based Urban Developments' (KBUD). This paper focuses on the urban design aspect of such large scale long term developments which has been given less importance in the planning literature. The paper discusses two important challenges related to land use design planning currently faced by planners of such specialized spaces. Firstly, Long term land use designs have become inefficient tools to guide development as they are constantly subjected to changing market forces. Second, design criteria for fostering interactive environments remains sketchy for such knowledge precincts. We discuss one possible design criteria with a primary aim to enhance 'knowledge interactions' between different participants and their relevance to progress of knowledge intensive communities. Using the unique criteria a new framework for a simple land use design model (LUDM-KBUD) is proposed using Agent Based Modelling (ABM) technique. Such a land use design model can help planners to develop physical planning guidelines in a continuous manner that would create a spatial design with the goal of maximizing knowledge/information interactions between participants. The research will provide planners with an alternative dynamic methodology to design such long term post-industrial mixed use developments. Future work in this direction and their ensuing implications on zoning practices of Knowledge Based Urban Developments (KBUD) and similar large scale urban developments will also be discussed. |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:arz:wpaper:eres2013_33&r=cmp |
| By: | Alfonso R. Sánchez (Universidad Pablo de Olavide de Sevilla) |
| Abstract: | This paper simulates the future performance of the Spanish pension system using a large OLG model. We compare the system in place after the 2011 pension reform to that emerging after the latest (2013) institutional changes. In particular, we explore the workings of the new indexing mechanism, linking pension payments to life expectancy and to the system’s aggregate flows of income and expenditure. We consider several alternative eco-demographic environments in our analysis and assess the welfare consequences for the different cohorts affected. Overall, the new automatic adjusting mechanism is broadly successful in its goal of stabilising the financial condition of the system. But the welfare costs imposed on some cohorts (e g. young workers at the beginning of the reform) is very heavy. |
| Keywords: | pension reform, automatic adjustment mechanism, population aging, Spain |
| JEL: | D58 H55 J11 |
| Date: | 2014–07 |
| URL: | http://d.repec.org/n?u=RePEc:bde:wpaper:1420&r=cmp |
| By: | Krieger, Patrick |
| Abstract: | Purpose – This paper argues that usual key indicators used in real estate management only show differences to a planned level. For this reason we consider the construction of behavioural key indicators supporting the human decision-making and learning process to improve decision-making in real estate management.Design / methodology / approach – We draw on findings in the area of dynamic decision-making. This viewpoint implicates the existence of a dynamic system on the property level thus employing an analytical case for showing the dynamics in the rental cash flow pattern. Findings – The dynamics in real estate cash flow pattern are driven by interdependencies and feedback processes between a set of variables. Thus, traditional instruments of managing the rental process are prone to erroneous perception about the ‘real-world’. Furthermore classical key indicators neither support the human decision-making process as well as the learning process for understanding such dynamic systems. A computational system is proposed to address support for both the dynamic decision-making and learning process.Research limitations / implications – (1) More research is needed on the validation of dynamic behaviour in cash flow pattern. (2) Validation on the performance of decision-making in the specific context of the rental process is needed. For this paper it is intended to raise questions for future research rather than delivering answers to this difficult and complex topic. Originality / value – The paper provides a new viewpoint on decision-making in real estate management by means of transforming task methodology in real estate management.Paper type – Viewpoint, Case studyArticle Classification: R10, G02, G17, C58 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:arz:wpaper:eres2014_105&r=cmp |