The Contribution of Edgeworth, Bootstrap and Monte Carlo Methods in Financial Data

Authors: Edlira Donefski, Tina Donefski, Lorenc Ekonomi

Abstract:

Edgeworth Approximation, Bootstrap and Monte Carlo Simulations have a considerable impact on the achieving certain results related to different problems taken into study. In our paper, we have treated a financial case related to the effect that have the components of a Cash-Flow of one of the most successful businesses in the world, as the financial activity, operational activity and investing activity to the cash and cash equivalents at the end of the three-months period. To have a better view of this case we have created a Vector Autoregression model, and after that we have generated the impulse responses in the terms of Asymptotic Analysis (Edgeworth Approximation), Monte Carlo Simulations and Residual Bootstrap based on the standard errors of every series created. The generated results consisted of the common tendencies for the three methods applied, that consequently verified the advantage of the three methods in the optimization of the model that contains many variants.

Keywords: Autoregression, Bootstrap, Edgeworth Expansion, Monte Carlo Method.

References:

[1] A. Bose, Edgeworth Correction by Bootstrap in Autoregressions, The Annals of Statistics, Vol.16, No.4 (1988), 1709-1722.
[2] A. Stuart and K. Ord, Distribution theory, vol.1. Oxford University press Inc., New York, USA, 1994, 74-162.
[3] C. Sims, Macroeconomics and Reality, Econometrica, 48 (1980), 1-48.
[4] D. Edgerton, and G. Shukur, Testing autocorrelation in a system perspective, Econometric Reviews, 18(1999), 343–386.
[5] F. Götze and C. Hipp, Asymptotic expansions for sums of weakly dependent random vectors, Z. Wahrsch. Verw.Gebiete, 64 (1983), 211-239.
[6] G.Casella and R. L. Berger, Statistical Inference, second edition. The Wadsworth Group, USA (2002), 83-87.
[7] H. Lütkepohl, Asymptotic distributions of impulse response functions and forecast error variance decompositions of vector autoregressive models. The Review Of Economics And Statistics, 72(1990), 116–125.
[8] H. Lütkepohl, Introduction to Multiple Time Series Analysis, New York: Springer-Verlag, 1991.
[9] H. Lütkepohl, New Introduction to Multiple Time Series Analysis, New York: Springer-Verlag, 2007.
[10] P. Hall, The Bootstrap and Edgeworth Expansion. Springer-Verlag, New York, USA, 1992.
[11] R. Basna, Edgeworth Expansion and Saddlepoint Approximation for Descrete Data with Applications in Chance Games. Linnéuniversitetet,2010.
[12] R. N. Bhattacharya and J. K. Ghosh, on the Validity of the Formal Edgeworth Expansion, The Annals of Statistics, Vol.6, No.2, (1978), 434-451
[13] R. W. Butler, Saddlepoint Approximation with Applications. Cambridge University Press, New York, USA (2007), 145.
[14] S. Johansen, Likelihood-based Inference in Cointegrated Vector Autoregressive Models, Oxford: Oxford University Press, 1995.A. H. Welsh, Asymptotically Efficient Estimation of the Sparsity Function at a Point, Statistics & Probability Letters, 6 (1988), 427-432.
[15] EViews Help: Multiple Equation Analysis
[16] http://fbemoodle.emu.edu.tr/pluginfile.php/47097/mod_resource/content/1/Workshop%204%20-%20Part%202%20-%20Advanced%20Time%20Series%20Econometrics%20with%20EViews.pdf