|
on Computational Economics |
|
Issue of 2013‒05‒24
three papers chosen by |
| By: | Andrew Feltenstein (Andrew Young School of Policy Studies, Georgia State University); Luciana Lopes; Janet Porras Mendoza (International Center for Public Policy. Andrew Young School of Policy Studies, Georgia State University); Sally Wallace (Andrew Young School of Policy Studies, Georgia State University) |
| Abstract: | Computational general equilibrium models (CGE) and micro-simulation models (MSM) each have their own sets of strengths and weaknesses. Both have been widely used for the analysis of fiscal policies in developing countries, and many attempts have been made to link the two models, thereby combining their relative strengths. We survey a broad literature that uses a variety of approaches to apply linked CGE and MSM models to analyze fiscal policies, in particular taxes and tariffs, in developing countries. We conclude that the “top down” approach, in which the aggregate outputs of the CGE model feed into the MSM, is the most commonly used. Nonetheless, a “bottom up” approach, in which the MSM generates estimated parameters, such as effective tax rates, which are then used as inputs to the CGE, may also be quite useful. As an example, we also develop both CGE and MSM models of Pakistan in order to indicate the relative uses of each model, although we have not at this time linked the two models. |
| Keywords: | Computable General Equilibrium, Micro-simulation, Fiscal Policy, Pakistan |
| Date: | 2013–04–07 |
| URL: | http://d.repec.org/n?u=RePEc:ays:ispwps:paper1309&r=cmp |
| By: | C. Alberto Dorantes (EGAP, Querétaro) |
| Abstract: | Several econometric software offer portfolio management tools for practitioners and researchers. For example, MatLab and R offer a great variety of tools for the simulation, optimization, and analysis of financial time series. Stata, together with Mata, offers powerful programming tools for the simulation, optimization, and analysis of financial data. However, related user commands are scarce. In this presentation, commands for online market data collection, data manipulation, and financial analysis for portfolio optimization are presented. Besides illustrating how these commands work, I will present the optimization algorithm used for portfolio optimization. Some of the commands include retornosyh. Based on the stockquote user command, this command retrieves online market data about stocks and indices (prices), computes returns—simple and continuously compounded—and then integrates the data into a formatted time-series dataset ready for further analysis. The following commands can be used to analyze data obtained with this command: gmvportws and gmvportwos, using series of continuously compounded returns, estimate the global minimum variance portfolio with and without short sales; and efrontier generates N portfolios in the efficient frontier and provides a graph showing the efficient frontier with and without short sales. The optimization algorithm was designed to estimate minimum variance portfolios according to Markowitz portfolio theory. When allowing for short sales, the minimum variance portfolio is analytically estimated based on Lagrange multipliers. When short sales are not allowed (nonnegative weights), there is no analytical solution, so the literature recommends using quadratic programming to find the minimum variance portfolio. Instead of following this numeric algorithm, I used an algorithm based on iterations and the analytical formula derived from Lagrange multipliers to estimate the minimum variance portfolio. Details and results will be described in the presentation. |
| Date: | 2013–05–13 |
| URL: | http://d.repec.org/n?u=RePEc:boc:msug13:08&r=cmp |
| By: | Anna Kowalska-Pyzalska; Katarzyna Maciejowska; Katarzyna Sznajd-Weron; Rafal Weron |
| Abstract: | Using an agent-based modeling approach we show how personal attributes, like conformity or indifference, impact the opinions of individual electricity consumers regarding switching to innovative dynamic tariff programs. We also examine the influence of advertising, discomfort of usage and the expectations of financial savings on opinion dynamics. Our main finding is that currently the adoption of dynamic electricity tariffs is virtually impossible due to the high level of indifference in today's societies. However, if in the future the indifference level is reduced, e.g., through educational programs that would make the customers more engaged in the topic, factors like tariff pricing schemes and intensity of advertising will became the focal point. |
| Keywords: | Dynamic pricing; Time-of-use tariff; Demand response; Diffusion of innovations; Agent-based model; Spinson |
| JEL: | C63 O33 Q48 Q55 |
| Date: | 2013–05–17 |
| URL: | http://d.repec.org/n?u=RePEc:wuu:wpaper:hsc1305&r=cmp |