nep-cmp New Economics Papers
on Computational Economics
Issue of 2015‒12‒20
eight papers chosen by

  1. Nonlinear time series and neural-network models of exchange rates between the US dollar and major currencies By Allen, D.E.; McAleer, M.J.; Peiris, S.; Singh, A.K.
  2. Bayesian Estimation of Agent-Based Models. By Jakob Grazzini; Matteo G. Richiardi; Mike Tsionas
  3. Deep Learning Stock Volatilities with Google Domestic Trends By Ruoxuan Xiong; Eric P. Nicholas; Yuan Shen
  4. Exact Present Solution with Consistent Future Approximation: A Gridless Algorithm to Solve Stochastic Dynamic Models By Wouter den Haan; Michal Kobielarz; Pontus Rendahl
  5. Calibrating the Dynamic Nelson-Siegel Model: A Practitioner Approach By Ibanez, Francisco
  6. A Simulation Analysis of the Longer-Term Effects of Immigration on Per Capita Income in an Aging Population By Frank T. Denton; Byron G. Spencer
  7. Climate policy and competitiveness: Policy guidance and quantitative evidence (Payne Institute Policy Brief) By Jared C. Carbone; Nicholas Rivers
  8. Informatics, Data Mining, Econometrics and Financial Economics: A Connection By Chang, C-L.; McAleer, M.J.; Wong, W-K.

  1. By: Allen, D.E.; McAleer, M.J.; Peiris, S.; Singh, A.K.
    Abstract: This paper features an analysis of major currency exchange rate movements in relation to the US dollar, as constituted in US dollar terms. Euro, British pound, Chinese yuan, and Japanese yen are modelled using a variety of non- linear models, including smooth transition regression models, logistic smooth transition regressions models, threshold autoregressive models, nonlinear autoregressive models, and additive nonlinear autoregressive models, plus Neural Network models. The results suggest that there is no dominating class of time series models, and the different currency pairs relationships with the US dollar are captured best by neural net regression models, over the ten year sample of daily exchange rate returns data, from August 2005 to August 2015.
    Keywords: non linear models, time series, non-parametric, smooth-transition regression models, neural networks, GMDH shell
    JEL: C45 C53 F3 G15
    Date: 2015–11–01
  2. By: Jakob Grazzini; Matteo G. Richiardi; Mike Tsionas
    Abstract: We consider Bayesian inference techniques for Agent-Based (AB) models, as an alternative to simulated minimum distance (SMD). We discuss the specicities of AB models with respect to models with exact aggregation results (as DSGE models), and how this impact estimation. Three computationally heavy steps are involved: (i) simulating the model, (ii) estimating the likelihood and (iii) sampling from the posterior distribution of the parameters. Computational complexity of AB models implies that ecient techniques have to be used with respect to points (ii) and (iii), possibly involving approximations. We rst discuss non-parametric (kernel density) estimation of the likelihood, coupled with Markov chain Monte Carlo sampling schemes. We then turn to parametric approximations of the likelihood, which can be derived by observing the distribution of the simulation outcomes around the statistical equilibria, or by assuming a specic form for the distribution of external deviations in the data. Finally, we introduce Approximate Bayesian Computation techniques for likelihood-free estimation. These allow embedding SMD methods in a Bayesian framework, and are particularly suited when robust estimation is needed. These techniques are tested, for the sake of comparison, in the same price discovery model used by Grazzini and Richiardi (2015) to illustrate SMD techniques.
    Date: 2015
  3. By: Ruoxuan Xiong; Eric P. Nicholas; Yuan Shen
    Abstract: We have applied the Long Short-Term Memory neural network to model S&P 500 volatilities incorporating Google domestic trends as indicators of the public mood and macroeconomic factors. In the 30% testing data, our Long Short-Term Memory model gives a mean absolute percentage error of 24.2%, outperforming linear Ridge/Lasso and autoregressive Garch benchmarks by at least 31%. This evaluation is done on the optimal observation and normalization scheme which maximizes the mutual information. Our preliminary investigation shows strong potential to better understand stock behaviors using deep learning neural network structures.
    Date: 2015–12
  4. By: Wouter den Haan (London School of Economics (LSE); Centre for Macroeconomics (CFM); Centre for Economic Policy Research (CEPR)); Michal Kobielarz (Tilburg University); Pontus Rendahl (Cambridge University; Centre for Macroeconomics (CFM))
    Abstract: This paper proposes an algorithm that finds model solutions at a particular point in the state space by solving a simple system of equations. The key step is to characterize future behavior with a Taylor series expansion of the current period's behavior around the contemporaneous values for the state variables. Since current decisions are solved from the original model equations, the solution incorporates nonlinearities and uncertainty. The algorithm is used to solve the model considered in Coeurdacier, Rey, and Winant (2011), which is a challenging model because it has no steady state and uncertainty is necessary to keep the model well behaved. We show that our algorithm can generate accurate solutions even when the model series are quite volatile. The solutions generated by the risky-steady-state algorithm proposed in Coeurdacier, Rey, and Winant (2011), in contrast, is shown to be not accurate.
    Keywords: Risky Steady State, Solution Methods
    JEL: C63 E10 E23 F41
    Date: 2015–12
  5. By: Ibanez, Francisco
    Abstract: Abstract The dynamic version of the Nelson-Siegel model has shown useful applications in the investment management industry. These applications go from forecasting the yield curve to portfolio risk management. Because of the complexity in the estimation of the parameters, some practitioners are unable to benefit from the uses of this model. In this note we present two approximations to estimate the time series of the model's factors. The first one has a more technical aim, focusing on the construction of a representative base to work, and uses a genetic algorithm to face the optimization problem. The second approximation has a practitioner spirit, focusing on the easiness of implementation. The results show that both approximations have good fitting for the U.S. Treasury bonds market.
    Keywords: Yield curve; Curve fitting; Calibration; Nelson-Siegel
    JEL: C51 C61 G12
    Date: 2015–12–14
  6. By: Frank T. Denton; Byron G. Spencer
    Abstract: Immigration is a possible instrument for offsetting longer-run adverse effects of population aging on per capita income. Our “laboratory” is a fictional country Alpha to which we assign demographic characteristics typical of a country experiencing population aging. Simulations indicate that a very high immigration rate with heavy concentration in younger working ages might be required to keep per capita income from declining. More rapid productivity growth would also offset population aging as would higher rates of labour participation of older people. Longer life expectancy, taken alone, would lower per capita real income, as would higher fertility rates.
    Keywords: immigration, per capita income, population aging, age structure, simulation
    JEL: J10 J11 J18
    Date: 2015–12
  7. By: Jared C. Carbone (Division of Economics and Business, Colorado School of Mines); Nicholas Rivers (University of Ottawa)
    Keywords: competitiveness, leakage, policy, carbon tax, climate change, computable general equilibrium
    JEL: C68 Q52 Q54
    Date: 2015–05
  8. By: Chang, C-L.; McAleer, M.J.; Wong, W-K.
    Abstract: This short communication reviews some of the literature in econometrics and financial economics that is related to informatics and data mining. We then discuss some of the research on econometrics and financial economics that could be extended to informatics and data mining beyond the existing areas in econometrics and financial economics.
    Keywords: econometrics, financial economics, informatics, data mining, theory, statistics
    JEL: C01 C81 C82 G14 L86 P34
    Date: 2015–11–01

General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.